Research Goal
Design one difficult but fair browsing challenge.
Web research • AI evaluation • Benchmark design
A legal-records benchmark, answered from law reports and an archived judgment paper.
Law reporting has its own conventions — parallel citations, Latin date lines, a Scottish series and an English one for the same decision — and a browsing agent has to read them correctly rather than trust a summary. This challenge describes a well-known appeal without naming it, and asks for the date judgment was handed down.
Frontier AI browsing systems can answer many questions by locating obvious keywords or matching information from highly ranked webpages. This project explores how to construct a research question that cannot be solved through a single search. The answer must remain short, stable and objectively verifiable, while the clues require the researcher to connect different people, dates, places, organisations, publications, events, records and quantities.
Design one difficult but fair browsing challenge.
A short, stable and objectively verifiable fact.
Every clue must be independently supported by reliable evidence.
A question, answer, clue set, source trail, validation log and JSON record.
AIBRC-003
Portfolio integrity note: Every claim on this page was checked against the cited source before publication. No source, quotation, locator or search result is invented; where a detail could not be verified directly it is described rather than quoted.
Begin with a stable fact that can be confirmed through a reliable primary or authoritative source. Record the exact wording, document title, issuing organisation and publication date.
Develop clues involving different fact types, including dates, people, places, organisations, works, events, records and quantities. Each clue must be independently checkable.
Run obvious, broad and targeted search queries. Record the results to identify accidental shortcuts, direct answer leakage and misleading pages.
Explain how each source supports a specific clue and how the clues logically combine to identify one answer.
Recheck every source, confirm answer uniqueness, document limitations and export the completed benchmark record in a consistent JSON format.
Each clue constrains the answer from a different direction. No single row is sufficient on its own, and every row can be checked against a source another researcher can reach.
Statuses are editable so the matrix can be worked through row by row. Scroll sideways on small screens.
| Clue ID | Fact type | Clue statement | Constraint | Supporting source | Exact locator | Verification status |
|---|---|---|---|---|---|---|
| C01 | Event | The dispute began when a woman drank part of a bottle of ginger beer in a Paisley cafe in August 1928 and found the decomposed remains of a snail in it. | One incident, dated by the judgment itself to 26 August 1928. | Law report of the appeal, Lord Buckmaster's opinion | Opening sentences of the first opinion | Verified |
| C02 | Person | The pursuer brought the action as a pauper against the manufacturer of the drink, not against the cafe owner who served it. | Named parties: M'Alister (or Donoghue), appellant; David Stevenson, respondent. | Session Cases report header | Case heading, No. 5 | Verified |
| C03 | Place | The case travelled from the Scottish Court of Session to the final appellate court sitting in London. | A Scottish appeal decided by the House of Lords. | BAILII record of the House of Lords decision | Court and citation line | Verified |
| C04 | Organisation | Five law lords sat on the appeal, and the judgment paper is held in a parliamentary record series. | Lords Buckmaster, Atkin, Tomlin, Thankerton and Macmillan; Parliamentary Archives reference HL/PO/JU/4/3/873. | BAILII transcript header | Line beginning "Die Jovis" and the list of Lords Present | Verified |
| C05 | Publication or work | The decision is reported in the Appeal Cases series at page 562 of the 1932 volume. | Citation [1932] AC 562, also reported as 1932 SC (HL) 31. | Law report reproduction of [1932] AC 562 | Report header, first page | Verified |
| C06 | Record or quantity | One of the five opinions set out a general test for the duty of care, framed around who is so closely and directly affected by an act. | Lord Atkin's opinion; distinguishes this appeal from the other opinions delivered. | Judgment text, Lord Atkin's opinion | Passage stating the general duty | Verified |
| C07 | Date | The House took time for consideration and delivered judgment on a Thursday in the spring of the following decade. | Dated in the report as 1932 May 26 and in the transcript as "Die Jovis, 26 Maii, 1932". | Session Cases report and BAILII transcript | Date lines in both records | Verified |
Each card records a source that was opened and read before it was cited. Excerpts are quoted only where the wording was read directly; otherwise the record describes what the source shows.
A strong benchmark documents the searches a capable researcher or browsing agent would naturally attempt, including the ones that exposed the answer. Leakage is recorded rather than hidden, and the question is rewritten around it.
| Search number | Exact search query | Search engine or database | Date searched | Top relevant results | Answer revealed? | Problem discovered | Action taken |
|---|---|---|---|---|---|---|---|
| S01 | Donoghue v Stevenson [1932] UKHL 100 decided 26 May 1932 House of Lords Lord Atkin neighbour principle | Web search | 4 September 2026 | BAILII transcript PDF; law report reproduction of [1932] AC 562; Scottish Council of Law Reporting case report; university copies of the judgment | Yes | The query names the case and the date, so every result restates the answer. | The published question never names the case, the parties or the citation; the solver must identify the appeal from the facts first. |
| S02 | snail ginger beer Paisley cafe 1928 House of Lords appeal judgment date | Web search | 4 September 2026 | Scottish Council of Law Reporting digital resources; Scottish Archives for Schools; general encyclopaedia coverage | Yes | The snail detail is famous enough that the case is identified in one hop, and several pages then give the date. | The question withholds the drink and the animal, describing only a contaminated bottled beverage bought for a customer by a companion. |
| S03 | 1932 SC (HL) 31 Session Cases report Scottish appeal five law lords | Web search | 4 September 2026 | Scottish Council of Law Reporting case report page; National Records of Scotland catalogue entry for the process papers | No | Reaches the Scottish reporting series without exposing the date on the results page itself. | Kept as the route used for evidence E03 and E06. |
Every candidate source is scored from 1 to 5 against five criteria before it is allowed into the evidence chain.
Who created or issued the information?
Is the source directly connected to the event or record?
Does it support the precise claim being made?
Is the information unlikely to change?
Can another researcher locate the same evidence?
Export a report that maps every clue to the source it rests on, records the verification status and access date of each source, and lists the searches that exposed the answer.
Completion
13 of 15 checks confirmed
87%
The complete challenge is delivered as a single validated JSON document.
aibrc-003-benchmark-record.json
{
"challenge_id": "AIBRC-003",
"title": "Landmark Judgment Benchmark",
"project_type": "independent_demonstration",
"research_domain": "Law Reports and Court Records",
"difficulty": "advanced",
"question": "A Scottish shopper became ill after drinking part of a bottled soft drink that a companion had bought for her in a Paisley cafe in 1928, and she sued the manufacturer rather than the seller. Five law lords finally decided her appeal in London. On what date was that judgment delivered?",
"accepted_answers": [
"26 May 1932"
],
"answer_type": "short_text",
"answer_normalisation": {
"case_sensitive": false,
"ignore_punctuation": true,
"accepted_variations": [
"26 May 1932",
"May 26, 1932",
"26/05/1932",
"1932-05-26"
]
},
"clues": [
{
"clue_id": "C01",
"fact_type": "event",
"statement": "The dispute began when a woman drank part of a bottle of ginger beer in a Paisley cafe in August 1928 and found the decomposed remains of a snail in it.",
"constraint": "One incident, dated by the judgment itself to 26 August 1928.",
"source": "Law report of the appeal, Lord Buckmaster's opinion",
"locator": "Opening sentences of the first opinion",
"status": "verified"
},
{
"clue_id": "C02",
"fact_type": "person",
"statement": "The pursuer brought the action as a pauper against the manufacturer of the drink, not against the cafe owner who served it.",
"constraint": "Named parties: M'Alister (or Donoghue), appellant; David Stevenson, respondent.",
"source": "Session Cases report header",
"locator": "Case heading, No. 5",
"status": "verified"
},
{
"clue_id": "C03",
"fact_type": "place",
"statement": "The case travelled from the Scottish Court of Session to the final appellate court sitting in London.",
"constraint": "A Scottish appeal decided by the House of Lords.",
"source": "BAILII record of the House of Lords decision",
"locator": "Court and citation line",
"status": "verified"
},
{
"clue_id": "C04",
"fact_type": "organisation",
"statement": "Five law lords sat on the appeal, and the judgment paper is held in a parliamentary record series.",
"constraint": "Lords Buckmaster, Atkin, Tomlin, Thankerton and Macmillan; Parliamentary Archives reference HL/PO/JU/4/3/873.",
"source": "BAILII transcript header",
"locator": "Line beginning \"Die Jovis\" and the list of Lords Present",
"status": "verified"
},
{
"clue_id": "C05",
"fact_type": "publication or work",
"statement": "The decision is reported in the Appeal Cases series at page 562 of the 1932 volume.",
"constraint": "Citation [1932] AC 562, also reported as 1932 SC (HL) 31.",
"source": "Law report reproduction of [1932] AC 562",
"locator": "Report header, first page",
"status": "verified"
},
{
"clue_id": "C06",
"fact_type": "record or quantity",
"statement": "One of the five opinions set out a general test for the duty of care, framed around who is so closely and directly affected by an act.",
"constraint": "Lord Atkin's opinion; distinguishes this appeal from the other opinions delivered.",
"source": "Judgment text, Lord Atkin's opinion",
"locator": "Passage stating the general duty",
"status": "verified"
},
{
"clue_id": "C07",
"fact_type": "date",
"statement": "The House took time for consideration and delivered judgment on a Thursday in the spring of the following decade.",
"constraint": "Dated in the report as 1932 May 26 and in the transcript as \"Die Jovis, 26 Maii, 1932\".",
"source": "Session Cases report and BAILII transcript",
"locator": "Date lines in both records",
"status": "verified"
}
],
"evidence": [
{
"evidence_id": "E01",
"claim_supported": "Supports C07 and the answer — the date judgment was delivered.",
"source_title": "Donoghue v Stevenson [1932] UKHL 100 (26 May 1932) — BAILII transcript",
"issuing_organisation": "British and Irish Legal Information Institute (BAILII)",
"source_type": "Institutional records",
"url": "http://www.bailii.org/uk/cases/UKHL/1932/100.html",
"publication_date": "Judgment delivered 26 May 1932",
"page_number": "n/a (transcript)",
"section": "Header, above the opinions",
"access_date": "4 September 2026",
"reliability": "High — transcript of the judgment itself",
"verification_status": "verified"
},
{
"evidence_id": "E02",
"claim_supported": "Corroborates C07 and supports C05 — the citation and the bench.",
"source_title": "Donoghue v Stevenson, [1932] AC 562 — law report reproduction (PDF)",
"issuing_organisation": "Università degli Studi di Trieste teaching repository",
"source_type": "PDF documents",
"url": "https://moodle2.units.it/pluginfile.php/441818/mod_resource/content/1/EN_Donoghue.pdf",
"publication_date": "Report of the 1932 judgment",
"page_number": "Page 1",
"section": "Report header",
"access_date": "4 September 2026",
"reliability": "Medium-high — faithful reproduction, but not the publisher's own copy",
"verification_status": "verified"
},
{
"evidence_id": "E03",
"claim_supported": "Supports C02, C05 and C07 from the Scottish report series.",
"source_title": "Donoghue v Stevenson — Case report, 1932 SC (HL) 31",
"issuing_organisation": "Scottish Council of Law Reporting",
"source_type": "Institutional records",
"url": "https://www.scottishlawreports.org.uk/resources/donoghue-v-stevenson/case-report/",
"publication_date": "Session Cases (House of Lords) 1932",
"page_number": "Report page 31",
"section": "Case heading, No. 5",
"access_date": "4 September 2026",
"reliability": "High — official Scottish law reporter",
"verification_status": "verified"
},
{
"evidence_id": "E04",
"claim_supported": "Supports C01 — the facts and the 1928 incident.",
"source_title": "Donoghue v Stevenson Digital Resources",
"issuing_organisation": "Scottish Council of Law Reporting",
"source_type": "Reference databases",
"url": "https://www.scottishlawreports.org.uk/resources/donoghue-v-stevenson/",
"publication_date": "Continuously maintained resource",
"page_number": "n/a (web resource)",
"section": "Introduction and appeal papers index",
"access_date": "4 September 2026",
"reliability": "High",
"verification_status": "verified"
},
{
"evidence_id": "E05",
"claim_supported": "Supports C01 and C02 — the cafe, the seller and the pursuer's claim.",
"source_title": "Snail in the Bottle — The Claimant's Case",
"issuing_organisation": "Scottish Archives for Schools (National Records of Scotland)",
"source_type": "Archived webpages",
"url": "https://www.scottisharchivesforschools.org/snailInTheBottle/theClaimant'sCase.asp",
"publication_date": "Educational resource drawing on Court of Session papers",
"page_number": "n/a (web page)",
"section": "The Claimant's Case",
"access_date": "4 September 2026",
"reliability": "Medium-high — national archive education resource",
"verification_status": "verified"
},
{
"evidence_id": "E06",
"claim_supported": "Supports C04 — the archived judgment paper reference.",
"source_title": "National Records of Scotland catalogue, Court of Session process CS252/2299",
"issuing_organisation": "National Records of Scotland",
"source_type": "Government records",
"url": "https://catalogue.nrscotland.gov.uk/nrsonlinecatalogue/details.aspx?reference=CS252%2F2299",
"publication_date": "Catalogue record",
"page_number": "n/a (catalogue entry)",
"section": "Single record",
"access_date": "4 September 2026",
"reliability": "High — national archive catalogue",
"verification_status": "verified"
}
],
"search_validation": [
{
"search_id": "S01",
"query": "Donoghue v Stevenson [1932] UKHL 100 decided 26 May 1932 House of Lords Lord Atkin neighbour principle",
"platform": "Web search",
"search_date": "4 September 2026",
"relevant_results": "BAILII transcript PDF; law report reproduction of [1932] AC 562; Scottish Council of Law Reporting case report; university copies of the judgment",
"answer_exposed": true,
"notes": "The query names the case and the date, so every result restates the answer. The published question never names the case, the parties or the citation; the solver must identify the appeal from the facts first."
},
{
"search_id": "S02",
"query": "snail ginger beer Paisley cafe 1928 House of Lords appeal judgment date",
"platform": "Web search",
"search_date": "4 September 2026",
"relevant_results": "Scottish Council of Law Reporting digital resources; Scottish Archives for Schools; general encyclopaedia coverage",
"answer_exposed": true,
"notes": "The snail detail is famous enough that the case is identified in one hop, and several pages then give the date. The question withholds the drink and the animal, describing only a contaminated bottled beverage bought for a customer by a companion."
},
{
"search_id": "S03",
"query": "1932 SC (HL) 31 Session Cases report Scottish appeal five law lords",
"platform": "Web search",
"search_date": "4 September 2026",
"relevant_results": "Scottish Council of Law Reporting case report page; National Records of Scotland catalogue entry for the process papers",
"answer_exposed": false,
"notes": "Reaches the Scottish reporting series without exposing the date on the results page itself. Kept as the route used for evidence E03 and E06."
}
],
"quality_checks": {
"The answer is short and objectively verifiable.": true,
"The answer is supported by an authoritative source.": true,
"The question has only one defensible answer.": true,
"No single clue directly exposes the answer.": false,
"Each clue is independently checkable.": true,
"The clue set spans multiple fact types.": true,
"The task requires multi-step reasoning.": true,
"Obvious searches have been documented.": true,
"Search-result snippets do not reveal the answer.": false,
"Sources are publicly accessible.": true,
"Exact pages, tables or sections are recorded.": true,
"The evidence chain can be reproduced.": true,
"The question avoids subjective interpretation.": true,
"The answer is stable over time.": true,
"No fabricated evidence is included.": true
},
"researcher": {
"name": "TMK Research Lab",
"portfolio": "https://theophilusmorkporkpor.lovable.app",
"email": "tkudjordjie@gmail.com"
}
}AI-generated statements are never treated as evidence. Every factual claim must be verified against an accessible external source.
This benchmark is complete: the answer is fixed, 6 evidence records are attached to specific clues, and 3 validation searches are documented, including those that exposed the answer. It demonstrates the ability to approach an unfamiliar subject systematically, separate evidence from assumption and document research so that another person can audit it.
A judgment has several competing dates: the incident in 1928, the hearings, the interlocutor in the Court of Session, and the day the House of Lords finally spoke. Only the last one answers the question.
Searching by the Session Cases citation alone reached the Scottish report but not a page that dated the decision plainly; the transcript header was needed.
The judgment transcript, whose header reads "Die Jovis, 26 Maii, 1932" and cites the Parliamentary Archives reference for the judgment paper.
Clue C07 pins the date to the delivery of judgment, and two independent reporting series print the same 26 May 1932.
Reporting 26 August 1928 (the incident), citing the earlier [1931] UKHL 3 parallel reference as if it were a separate decision, or misreading the Latin date line.
One supporting copy of the Appeal Cases report is a university teaching PDF rather than the publisher's own edition; the transcript and the Scottish report carry the weight.
Consult the Parliamentary Archives judgment paper HL/PO/JU/4/3/873 directly and add archived copies of each cited page.
I am available for web research, fact-checking, AI evaluation, benchmark construction, source verification and structured research documentation.
TMK Research Labtkudjordjie@gmail.comhttps://theophilusmorkporkpor.lovable.apphttps://www.linkedin.com/in/theophilus-kudzordzi