{ "id": "AIBD-0351", "title": "GiveDirectly's hybrid MobileAid program in Malawi may have enrolled 35% of recipients by phone-metadata machine learning and halved enrollment cost", "summary": "GiveDirectly posts (2024–2025) describe phone-metadata AI/ML targeting in Malawi: 35% of recipients enrolled via phone data, cost per enrolee halved; no date, count or independent source.", "date_observed": "2025-09-04", "date_basis": "unassessed", "date_precision": "unassessed", "benefit_status": null, "source_collection": null, "source_entry": null, "account": "GiveDirectly's MobileAid approach uses machine learning (ML) on mobile-phone usage patterns to identify people in need and pay them by mobile money. Its 4 September 2025 post (Vera Lummis) says it has evaluated the method across four countries since 2021 (Togo, DRC, Malawi and Bangladesh). In Malawi it ran a hybrid program identifying and enrolling 35% of recipients via phone metadata and 65% in person. It says this halved the cost of identifying and enrolling each additional person compared with a fully in-person operation. It is working with the Government of Malawi to fill gaps in the urban social registry with phone metadata. A study with the Government of Malawi, U.C. Berkeley and U.C. San Diego is expected to report in 2026. It asks whether phone metadata can flag people whose poverty deepens after a shock. Its 27 January 2025 post (Nirali Desai and Vera Lummis) describes interviews with 67 recipients across Malawi and Kenya in February and October 2024. The interviews, naming Mzuzu, asked how to explain AI/ML targeting. It finds scenario-based explanations most understandable. No cited source gives the Malawi program's dates, location, number of recipients or an independent evaluation.", "category": "general benefit", "unexpected": false, "purpose": "Identify people in poverty from mobile-phone metadata with machine learning so that cash can be delivered faster and more cheaply. The claimed benefit (cheaper, faster enrollment) is that purpose.", "system": "GiveDirectly MobileAid: machine-learning poverty targeting from mobile-phone metadata", "model_version": null, "developer": "GiveDirectly (MobileAid), with research partners; for the Malawi study the Government of Malawi, U.C. Berkeley and U.C. San Diego", "deployer": "GiveDirectly, with the Government of Malawi for the urban social registry work", "location": "Malawi (the explain-AI post names Mzuzu)", "domain": "Humanitarian cash transfers / poverty targeting", "segment": "Non-governmental cash-transfer organization working with a national government", "benefit_type": "Improved Organizational Efficiency", "taxonomy_note": null, "registry_ids": {}, "ratings": [ { "group": "personal", "depth": null, "reason": "No dated event, named recipient or outcome for Malawi is given in any cited source." }, { "group": "organizational", "depth": null, "reason": "The halved enrollment cost is the maker's own undated statement without figures or an independent evaluation." }, { "group": "societal", "depth": null, "reason": "No population-level outcome is documented for Malawi." } ], "ai_role": { "status": "assessed", "statement": "Only the maker's account is among the sources. GiveDirectly says machine learning on phone metadata identified and enrolled 35% of recipients in a Malawi hybrid program. It says this halved the cost per additional enrolee. It gives no date, place, recipient count or accuracy for Malawi. The Government of Malawi, U.C. Berkeley and U.C. San Diego study is described as pending (findings expected in 2026). Likewise, the explain-AI page names no deployment, date or outcome." }, "evidence": [ { "kind": "press release", "citation": "Lummis, V. (2025). Five things we've learned using mobile data and AI/ML to identify people in need. GiveDirectly blog, 4 September 2025.", "title": "Five things we've learned using mobile data and AI/ML to identify people in need", "url": "https://www.givedirectly.org/mobileaid", "date": "2025-09-04", "written_by": "makers", "quote": "So in Malawi, we ran a hybrid program: we identified and enrolled 35% of recipients via phone metadata and the other 65% in person." }, { "kind": "press release", "citation": "Lummis, V. (2025). Five things we've learned using mobile data and AI/ML to identify people in need. GiveDirectly blog, 4 September 2025 (cost statement).", "title": "Five things we've learned using mobile data and AI/ML to identify people in need (cost statement)", "url": "https://www.givedirectly.org/mobileaid", "date": "2025-09-04", "written_by": "makers", "quote": "This combined approach still cut the cost of identifying and enrolling each additional person in half, compared to a fully in-person operation." }, { "kind": "press release", "citation": "Desai, N. & Lummis, V. (2025). How to explain AI/ML across the digital divide. GiveDirectly blog, 27 January 2025.", "title": "Explain AI/ML systems to people in poverty using real-world scenarios", "url": "https://www.givedirectly.org/explain_ai", "date": "2025-01-27", "written_by": "makers", "quote": "we interviewed 67 GiveDirectly recipients across Malawi and Kenya in February and October 2024" }, { "kind": "press release", "citation": "Lummis, V., Luk, S. & Ramprasad, S. (2024). GiveDirectly's approach to responsible AI/ML. GiveDirectly blog, 8 July 2024.", "title": "GiveDirectly's approach to responsible AI/ML", "url": "https://www.givedirectly.org/ai-framework", "date": "2024-07-08", "written_by": "makers", "quote": "interviews with GiveDirectly recipients in Malawi and Kenya" } ], "replication": { "produced_by": [ "Access to mobile network operator metadata for the target population under a data agreement", "A machine-learning model relating phone usage patterns to poverty, trained on survey data", "Mobile-money rails to enroll and pay identified people", "An in-person survey component (65% of enrollments) alongside the phone-metadata component" ], "must_hold": [ "Telecom partnership and data-sharing agreement in place (GiveDirectly cites 6–18 months set-up)", "Phone ownership and usage among the poor population", "Recipients' understanding and consent, which GiveDirectly found requires scenario-based explanation", "Model accuracy acceptable for the program (GiveDirectly reports phone metadata below in-person proxy-means tests)" ], "echoes": [ { "date": null, "setting": "Togo Novissi COVID-19 cash program (precedent, see AIBD-0093): GiveDirectly states the government paid over 140,000 people within weeks", "model_version": null, "outcome": "recurred", "source": "https://www.givedirectly.org/mobileaid" }, { "date": "2023-12-21", "setting": "Bangladesh: GiveDirectly states recipients were paid as quickly as 3 days after registration using machine-learning eligibility analysis with CEGA", "model_version": null, "outcome": "recurred", "source": "https://www.givedirectly.org/tech-2023" }, { "date": "2025-09-04", "setting": "Kenya: enrollment in flood- and drought-prone areas with Safaricom processing phone metadata in-house, described as ongoing", "model_version": null, "outcome": "unknown", "source": "https://www.givedirectly.org/mobileaid" } ] }, "paired_incident": null, "headline": "GiveDirectly's MobileAid may have halved enrollment cost in Malawi", "metric": "35% of recipients enrolled via phone data, cost per enrolee halved", "comparator": { "against": "a fully in-person enrollment operation", "before": null, "after": "cut the cost of identifying and enrolling each additional person in half", "basis": "measured", "quote": "This combined approach still cut the cost of identifying and enrolling each additional person in half, compared to a fully in-person operation.", "url": "https://www.givedirectly.org/mobileaid" }, "later": { "status": "unknown", "date": null, "note": "GiveDirectly says a study with the Government of Malawi, U.C. Berkeley and U.C. San Diego is expected to report in 2026; nothing dated after 4 September 2025 was found in the cited sources.", "quote": null, "url": null }, "value": null, "developer_country": "United States", "deployer_country": "United States", "unrated_because": null, "brief_notes": null, "brief": null, "revisions": [ { "date": "2026-10-20", "change": "Record drafted from the cited sources.", "source": null, "editor": "AIBD" }, { "date": "2026-10-21", "change": "Headline reworded to state the event without naming who claims it.", "source": null, "editor": "AIBD" }, { "date": "2026-10-21", "change": "Title reworded to state the event without naming who claims it.", "source": null, "editor": "AIBD" }, { "date": "2026-10-21", "change": "Added what the benefit is measured against from the cited sources.", "source": "https://www.givedirectly.org/mobileaid", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Country of the developer and the deployer added from the organization's own site or a registry.", "source": "https://projects.propublica.org/nonprofits/organizations/271661997; https://projects.propublica.org/nonprofits/organizations/271661997", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Prose pass: account, purpose, the replication conditions, the system's part rewritten in shorter sentences with abbreviations expanded; facts, figures and quotations unchanged.", "source": null, "editor": "AIBD" }, { "date": "2026-10-21", "change": "Notes about which sources could be read were reworded for readers; facts, figures and quotations unchanged.", "source": null, "editor": "AIBD" } ], "confidence": "claimed", "shade": "low", "claim_status": "The makers' claim", "claim_basis": "Sources: three press releases from the makers. The cited evidence includes no independent source and no finding by a regulator or court.", "profile": { "case_id": "AIBD-0351", "catalog_version": "1.0.0", "source_fingerprint": "0f3426e3243fafb4484c0c9f94cfde3f8aa82ff86c675d046679b52e8eff7a1a", "collection_refs": [ { "collection_id": "aibd", "version": "snapshot-8bf7cfec9cb185aa" }, { "collection_id": "aibd", "version": "2026-10-25" }, { "collection_id": "viator", "version": "2026-10-25" } ], "source_ids": [ "AIBD-0351-source-0001", "AIBD-0351-source-0002", "AIBD-0351-source-0003", "AIBD-0351-source-0004" ], "families": { "strategic assessments": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "record_ids": [] }, "evaluations": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "record_ids": [] }, "monetary observations": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "record_ids": [] }, "institutional decisions": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "record_ids": [] } }, "fields": { "actors[]": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "developer": "GiveDirectly (MobileAid), with research partners; for the Malawi study the Government of Malawi, U.C. Berkeley and U.C. San Diego", "deployer": "GiveDirectly, with the Government of Malawi for the urban social registry work" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "developer", "deployer" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "actors[].ownership_control": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "ai_contribution": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "ai_role": { "status": "assessed", "statement": "Only the maker's account is among the sources. GiveDirectly says machine learning on phone metadata identified and enrolled 35% of recipients in a Malawi hybrid program. It says this halved the cost per additional enrolee. It gives no date, place, recipient count or accuracy for Malawi. The Government of Malawi, U.C. Berkeley and U.C. San Diego study is described as pending (findings expected in 2026). Likewise, the explain-AI page names no deployment, date or outcome." } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ai_role" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "ai_relationship": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "category": "general benefit", "ai_role": { "status": "assessed", "statement": "Only the maker's account is among the sources. GiveDirectly says machine learning on phone metadata identified and enrolled 35% of recipients in a Malawi hybrid program. It says this halved the cost per additional enrolee. It gives no date, place, recipient count or accuracy for Malawi. The Government of Malawi, U.C. Berkeley and U.C. San Diego study is described as pending (findings expected in 2026). Likewise, the explain-AI page names no deployment, date or outcome." } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ai_role", "category" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "assessments[].alternatives": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "assessments[].assessor_date": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "assessments[].depth": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "assessments[].display_confidence": { "state": "value_present", "value": { "kind": "structured", "data": { "confidence": "low", "evidence_word": "claimed", "claim_status": "The makers' claim", "basis": "Sources: three press releases from the makers. The cited evidence includes no independent source and no finding by a regulator or court.", "interpretation": "Existing display rule preserved; the richer outcome-specific confidence question still needs review." } }, "provenance": { "kind": "curatorial", "basis": "Derived from the existing case display properties; not a new evidentiary judgment.", "record_fields": [], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "assessments[].rationale": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "ratings": [ { "group": "personal", "reason": "No dated event, named recipient or outcome for Malawi is given in any cited source." }, { "group": "organizational", "reason": "The halved enrollment cost is the maker's own undated statement without figures or an independent evaluation." }, { "group": "societal", "reason": "No population-level outcome is documented for Malawi." } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ratings" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "assessments[].rubric_version": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "assessments[].status_basis": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "assessments[].target": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "ratings": [ { "group": "personal", "reason": "No dated event, named recipient or outcome for Malawi is given in any cited source." }, { "group": "organizational", "reason": "The halved enrollment cost is the maker's own undated statement without figures or an independent evaluation." }, { "group": "societal", "reason": "No population-level outcome is documented for Malawi." } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ratings" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "availability": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "brief": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "case_kind": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "case_links[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "claims[].evidence_links": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "claims[].evidence_strength": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "claims[].missingness": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "claims[].proposition": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "cross_border_flows[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "decisions[].action_outcome": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "institutional decisions" }, "decisions[].grounds": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "institutional decisions" }, "decisions[].institution_task": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "institutional decisions" }, "decisions[].success_criterion": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "institutional decisions" }, "deployment_stage": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "economics[].amount": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].costs[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].counterfactual": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].currency_price_basis": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].gross_net": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].measure_type": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].period_unit": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].realization": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].recipient_payer": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].return_payback": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].valuation_method": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "evaluations[].ai_uplift": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].comparator": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].design": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].effect": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].exposure": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].id": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].measurement_method": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].metric": { "state": "not_assessed", "provenance": { "kind": "legacy_record", "basis": "Mapped from the existing case description without a new external source review.", "record_fields": [ "metric" ], "source_refs": [] }, "review_state": "not_assessed", "coverage": "partial", "family": "evaluations", "display": { "answer": "35% of recipients enrolled via phone data, cost per enrolee halved", "editor_note": "" } }, "evaluations[].reproduction_materials": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].setting_sample": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].uncertainty_limitations": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "external_references[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "follow_up[]": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "later": { "status": "unknown", "note": "GiveDirectly says a study with the Government of Malawi, U.C. Berkeley and U.C. San Diego is expected to report in 2026; nothing dated after 4 September 2025 was found in the cited sources." } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "later" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "human_oversight": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "id": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "id": "AIBD-0351" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "id" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "infrastructure_dependencies[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].beneficiaries": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].beneficiary_count": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].benefit_type": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "benefit_type": "Improved Organizational Efficiency" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "benefit_type" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "outcomes[].claim": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "account": "GiveDirectly's MobileAid approach uses machine learning (ML) on mobile-phone usage patterns to identify people in need and pay them by mobile money. Its 4 September 2025 post (Vera Lummis) says it has evaluated the method across four countries since 2021 (Togo, DRC, Malawi and Bangladesh). In Malawi it ran a hybrid program identifying and enrolling 35% of recipients via phone metadata and 65% in person. It says this halved the cost of identifying and enrolling each additional person compared with a fully in-person operation. It is working with the Government of Malawi to fill gaps in the urban social registry with phone metadata. A study with the Government of Malawi, U.C. Berkeley and U.C. San Diego is expected to report in 2026. It asks whether phone metadata can flag people whose poverty deepens after a shock. Its 27 January 2025 post (Nirali Desai and Vera Lummis) describes interviews with 67 recipients across Malawi and Kenya in February and October 2024. The interviews, naming Mzuzu, asked how to explain AI/ML targeting. It finds scenario-based explanations most understandable. No cited source gives the Malawi program's dates, location, number of recipients or an independent evaluation.", "benefit_type": "Improved Organizational Efficiency" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "account", "benefit_type" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "outcomes[].distribution": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].duration": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].id": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].stakeholder_group": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "ratings": [ { "group": "personal", "reason": "No dated event, named recipient or outcome for Malawi is given in any cited source." }, { "group": "organizational", "reason": "The halved enrollment cost is the maker's own undated statement without figures or an independent evaluation." }, { "group": "societal", "reason": "No population-level outcome is documented for Malawi." } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ratings" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "outcomes[].status": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].tradeoffs": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].unexpected": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "unexpected": false, "purpose": "Identify people in poverty from mobile-phone metadata with machine learning so that cash can be delivered faster and more cheaply. The claimed benefit (cheaper, faster enrollment) is that purpose." } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "unexpected", "purpose" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "places[]": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "location": "Malawi (the explain-AI post names Mzuzu)", "developer_country": "United States", "deployer_country": "United States" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "location", "developer_country", "deployer_country" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "purpose": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "purpose": "Identify people in poverty from mobile-phone metadata with machine learning so that cash can be delivered faster and more cheaply. The claimed benefit (cheaper, faster enrollment) is that purpose." } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "purpose" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "record_dates": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "date_observed": "2025-09-04", "revisions": [ { "date": "2026-10-20", "change": "Record drafted from the cited sources.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Headline reworded to state the event without naming who claims it.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Title reworded to state the event without naming who claims it.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Added what the benefit is measured against from the cited sources.", "source": "https://www.givedirectly.org/mobileaid", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Country of the developer and the deployer added from the organization's own site or a registry.", "source": "https://projects.propublica.org/nonprofits/organizations/271661997; https://projects.propublica.org/nonprofits/organizations/271661997", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Prose pass: account, purpose, the replication conditions, the system's part rewritten in shorter sentences with abbreviations expanded; facts, figures and quotations unchanged.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Notes about which sources could be read were reworded for readers; facts, figures and quotations unchanged.", "editor": "AIBD" } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "date_observed", "revisions" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "replication.attempts[]": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "replication": { "produced_by": [ "Access to mobile network operator metadata for the target population under a data agreement", "A machine-learning model relating phone usage patterns to poverty, trained on survey data", "Mobile-money rails to enroll and pay identified people", "An in-person survey component (65% of enrollments) alongside the phone-metadata component" ], "must_hold": [ "Telecom partnership and data-sharing agreement in place (GiveDirectly cites 6–18 months set-up)", "Phone ownership and usage among the poor population", "Recipients' understanding and consent, which GiveDirectly found requires scenario-based explanation", "Model accuracy acceptable for the program (GiveDirectly reports phone metadata below in-person proxy-means tests)" ], "echoes": [ { "setting": "Togo Novissi COVID-19 cash program (precedent, see AIBD-0093): GiveDirectly states the government paid over 140,000 people within weeks", "outcome": "recurred", "source": "https://www.givedirectly.org/mobileaid" }, { "date": "2023-12-21", "setting": "Bangladesh: GiveDirectly states recipients were paid as quickly as 3 days after registration using machine-learning eligibility analysis with CEGA", "outcome": "recurred", "source": "https://www.givedirectly.org/tech-2023" }, { "date": "2025-09-04", "setting": "Kenya: enrollment in flood- and drought-prone areas with Safaricom processing phone metadata in-house, described as ongoing", "outcome": "unknown", "source": "https://www.givedirectly.org/mobileaid" } ] } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "replication" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "replication.conditions": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "replication": { "produced_by": [ "Access to mobile network operator metadata for the target population under a data agreement", "A machine-learning model relating phone usage patterns to poverty, trained on survey data", "Mobile-money rails to enroll and pay identified people", "An in-person survey component (65% of enrollments) alongside the phone-metadata component" ], "must_hold": [ "Telecom partnership and data-sharing agreement in place (GiveDirectly cites 6–18 months set-up)", "Phone ownership and usage among the poor population", "Recipients' understanding and consent, which GiveDirectly found requires scenario-based explanation", "Model accuracy acceptable for the program (GiveDirectly reports phone metadata below in-person proxy-means tests)" ], "echoes": [ { "setting": "Togo Novissi COVID-19 cash program (precedent, see AIBD-0093): GiveDirectly states the government paid over 140,000 people within weeks", "outcome": "recurred", "source": "https://www.givedirectly.org/mobileaid" }, { "date": "2023-12-21", "setting": "Bangladesh: GiveDirectly states recipients were paid as quickly as 3 days after registration using machine-learning eligibility analysis with CEGA", "outcome": "recurred", "source": "https://www.givedirectly.org/tech-2023" }, { "date": "2025-09-04", "setting": "Kenya: enrollment in flood- and drought-prone areas with Safaricom processing phone metadata in-house, described as ongoing", "outcome": "unknown", "source": "https://www.givedirectly.org/mobileaid" } ] } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "replication" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "revision_history[]": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "revisions": [ { "date": "2026-10-20", "change": "Record drafted from the cited sources.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Headline reworded to state the event without naming who claims it.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Title reworded to state the event without naming who claims it.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Added what the benefit is measured against from the cited sources.", "source": "https://www.givedirectly.org/mobileaid", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Country of the developer and the deployer added from the organization's own site or a registry.", "source": "https://projects.propublica.org/nonprofits/organizations/271661997; https://projects.propublica.org/nonprofits/organizations/271661997", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Prose pass: account, purpose, the replication conditions, the system's part rewritten in shorter sentences with abbreviations expanded; facts, figures and quotations unchanged.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Notes about which sources could be read were reworded for readers; facts, figures and quotations unchanged.", "editor": "AIBD" } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "revisions" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "sector": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "domain": "Humanitarian cash transfers / poverty targeting", "segment": "Non-governmental cash-transfer organization working with a national government" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "domain", "segment" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "sources[].access_version": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "sources[].citation_locator": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "evidence": [ { "kind": "press release", "citation": "Lummis, V. (2025). Five things we've learned using mobile data and AI/ML to identify people in need. GiveDirectly blog, 4 September 2025.", "title": "Five things we've learned using mobile data and AI/ML to identify people in need", "url": "https://www.givedirectly.org/mobileaid", "date": "2025-09-04", "written_by": "makers", "quote": "So in Malawi, we ran a hybrid program: we identified and enrolled 35% of recipients via phone metadata and the other 65% in person." }, { "kind": "press release", "citation": "Lummis, V. (2025). Five things we've learned using mobile data and AI/ML to identify people in need. GiveDirectly blog, 4 September 2025 (cost statement).", "title": "Five things we've learned using mobile data and AI/ML to identify people in need (cost statement)", "url": "https://www.givedirectly.org/mobileaid", "date": "2025-09-04", "written_by": "makers", "quote": "This combined approach still cut the cost of identifying and enrolling each additional person in half, compared to a fully in-person operation." }, { "kind": "press release", "citation": "Desai, N. & Lummis, V. (2025). How to explain AI/ML across the digital divide. GiveDirectly blog, 27 January 2025.", "title": "Explain AI/ML systems to people in poverty using real-world scenarios", "url": "https://www.givedirectly.org/explain_ai", "date": "2025-01-27", "written_by": "makers", "quote": "we interviewed 67 GiveDirectly recipients across Malawi and Kenya in February and October 2024" }, { "kind": "press release", "citation": "Lummis, V., Luk, S. & Ramprasad, S. (2024). GiveDirectly's approach to responsible AI/ML. GiveDirectly blog, 8 July 2024.", "title": "GiveDirectly's approach to responsible AI/ML", "url": "https://www.givedirectly.org/ai-framework", "date": "2024-07-08", "written_by": "makers", "quote": "interviews with GiveDirectly recipients in Malawi and Kenya" } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "evidence" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "sources[].id": { "state": "value_present", "value": { "kind": "identifiers", "identifiers": [ "AIBD-0351-source-0001", "AIBD-0351-source-0002", "AIBD-0351-source-0003", "AIBD-0351-source-0004" ], "role": "Existing evidence items, in their original order" }, "provenance": { "kind": "curatorial", "basis": "Stable within the fingerprinted record version; no source content or source review has been added.", "record_fields": [], "source_refs": [] }, "review_state": "record_mapped", "coverage": "complete" }, "sources[].language": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "sources[].relationship": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "evidence": [ { "kind": "press release", "citation": "Lummis, V. (2025). Five things we've learned using mobile data and AI/ML to identify people in need. GiveDirectly blog, 4 September 2025.", "title": "Five things we've learned using mobile data and AI/ML to identify people in need", "url": "https://www.givedirectly.org/mobileaid", "date": "2025-09-04", "written_by": "makers", "quote": "So in Malawi, we ran a hybrid program: we identified and enrolled 35% of recipients via phone metadata and the other 65% in person." }, { "kind": "press release", "citation": "Lummis, V. (2025). Five things we've learned using mobile data and AI/ML to identify people in need. GiveDirectly blog, 4 September 2025 (cost statement).", "title": "Five things we've learned using mobile data and AI/ML to identify people in need (cost statement)", "url": "https://www.givedirectly.org/mobileaid", "date": "2025-09-04", "written_by": "makers", "quote": "This combined approach still cut the cost of identifying and enrolling each additional person in half, compared to a fully in-person operation." }, { "kind": "press release", "citation": "Desai, N. & Lummis, V. (2025). How to explain AI/ML across the digital divide. GiveDirectly blog, 27 January 2025.", "title": "Explain AI/ML systems to people in poverty using real-world scenarios", "url": "https://www.givedirectly.org/explain_ai", "date": "2025-01-27", "written_by": "makers", "quote": "we interviewed 67 GiveDirectly recipients across Malawi and Kenya in February and October 2024" }, { "kind": "press release", "citation": "Lummis, V., Luk, S. & Ramprasad, S. (2024). GiveDirectly's approach to responsible AI/ML. GiveDirectly blog, 8 July 2024.", "title": "GiveDirectly's approach to responsible AI/ML", "url": "https://www.givedirectly.org/ai-framework", "date": "2024-07-08", "written_by": "makers", "quote": "interviews with GiveDirectly recipients in Malawi and Kenya" } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "evidence" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "sources[].upstream_sources": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "strategic_assessments[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "strategic assessments" }, "summary": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "summary": "GiveDirectly posts (2024–2025) describe phone-metadata AI/ML targeting in Malawi: 35% of recipients enrolled via phone data, cost per enrolee halved; no date, count or independent source.", "account": "GiveDirectly's MobileAid approach uses machine learning (ML) on mobile-phone usage patterns to identify people in need and pay them by mobile money. Its 4 September 2025 post (Vera Lummis) says it has evaluated the method across four countries since 2021 (Togo, DRC, Malawi and Bangladesh). In Malawi it ran a hybrid program identifying and enrolling 35% of recipients via phone metadata and 65% in person. It says this halved the cost of identifying and enrolling each additional person compared with a fully in-person operation. It is working with the Government of Malawi to fill gaps in the urban social registry with phone metadata. A study with the Government of Malawi, U.C. Berkeley and U.C. San Diego is expected to report in 2026. It asks whether phone metadata can flag people whose poverty deepens after a shock. Its 27 January 2025 post (Nirali Desai and Vera Lummis) describes interviews with 67 recipients across Malawi and Kenya in February and October 2024. The interviews, naming Mzuzu, asked how to explain AI/ML targeting. It finds scenario-based explanations most understandable. No cited source gives the Malawi program's dates, location, number of recipients or an independent evaluation." } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "summary", "account" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "systems[].autonomy_permissions": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "systems[].component_role": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "systems[].deployment_configuration": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "systems[].inputs_outputs": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "systems[].model_name": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "systems[].model_version": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "systems[].product_name": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "system": "GiveDirectly MobileAid: machine-learning poverty targeting from mobile-phone metadata" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "system" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "systems[].provider": { "state": "value_present", "value": { "kind": "structured", "data": { "application_developer": "GiveDirectly (MobileAid), with research partners; for the Malawi study the Government of Malawi, U.C. Berkeley and U.C. San Diego" } }, "provenance": { "kind": "legacy_record", "basis": "Mapped from the existing case description without a new external source review.", "record_fields": [ "developer" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "taxonomy_crosswalks[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "title": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "title": "GiveDirectly's hybrid MobileAid program in Malawi may have enrolled 35% of recipients by phone-metadata machine learning and halved enrollment cost", "headline": "GiveDirectly's MobileAid may have halved enrollment cost in Malawi" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "title", "headline" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "version": { "state": "value_present", "value": { "kind": "identifiers", "identifiers": [ "0f3426e3243fafb4484c0c9f94cfde3f8aa82ff86c675d046679b52e8eff7a1a" ], "role": "SHA-256 of normalized original case data" }, "provenance": { "kind": "curatorial", "basis": "Deterministic content version assigned by the local migration.", "record_fields": [], "source_refs": [] }, "review_state": "record_mapped", "coverage": "complete" }, "workflow": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "account": "GiveDirectly's MobileAid approach uses machine learning (ML) on mobile-phone usage patterns to identify people in need and pay them by mobile money. Its 4 September 2025 post (Vera Lummis) says it has evaluated the method across four countries since 2021 (Togo, DRC, Malawi and Bangladesh). In Malawi it ran a hybrid program identifying and enrolling 35% of recipients via phone metadata and 65% in person. It says this halved the cost of identifying and enrolling each additional person compared with a fully in-person operation. It is working with the Government of Malawi to fill gaps in the urban social registry with phone metadata. A study with the Government of Malawi, U.C. Berkeley and U.C. San Diego is expected to report in 2026. It asks whether phone metadata can flag people whose poverty deepens after a shock. Its 27 January 2025 post (Nirali Desai and Vera Lummis) describes interviews with 67 recipients across Malawi and Kenya in February and October 2024. The interviews, naming Mzuzu, asked how to explain AI/ML targeting. It finds scenario-based explanations most understandable. No cited source gives the Malawi program's dates, location, number of recipients or an independent evaluation.", "replication": { "produced_by": [ "Access to mobile network operator metadata for the target population under a data agreement", "A machine-learning model relating phone usage patterns to poverty, trained on survey data", "Mobile-money rails to enroll and pay identified people", "An in-person survey component (65% of enrollments) alongside the phone-metadata component" ], "must_hold": [ "Telecom partnership and data-sharing agreement in place (GiveDirectly cites 6–18 months set-up)", "Phone ownership and usage among the poor population", "Recipients' understanding and consent, which GiveDirectly found requires scenario-based explanation", "Model accuracy acceptable for the program (GiveDirectly reports phone metadata below in-person proxy-means tests)" ], "echoes": [ { "setting": "Togo Novissi COVID-19 cash program (precedent, see AIBD-0093): GiveDirectly states the government paid over 140,000 people within weeks", "outcome": "recurred", "source": "https://www.givedirectly.org/mobileaid" }, { "date": "2023-12-21", "setting": "Bangladesh: GiveDirectly states recipients were paid as quickly as 3 days after registration using machine-learning eligibility analysis with CEGA", "outcome": "recurred", "source": "https://www.givedirectly.org/tech-2023" }, { "date": "2025-09-04", "setting": "Kenya: enrollment in flood- and drought-prone areas with Safaricom processing phone metadata in-house, described as ongoing", "outcome": "unknown", "source": "https://www.givedirectly.org/mobileaid" } ] } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "account", "replication" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" } }, "summary": { "total": 91, "addressed": 91, "known": 28, "reviewed": 0, "record_mapped": 28, "not_assessed": 63, "complete": false, "by_state": { "value_present": 28, "not_reported": 0, "unresolved": 0, "not_assessed": 63, "not_applicable": 0, "withheld": 0, "source_unavailable": 0 } } } }