Datasets:
prompt stringlengths 60 2.89k | chosen stringlengths 151 5.74k | rejected stringlengths 15 5.74k | source stringclasses 54
values | defect stringclasses 16
values | negatives stringclasses 3
values |
|---|---|---|---|---|---|
A food import firm 'Al-Mawarid General Trading LLC' (UAE, trade license 12345); list has 'Al Mawarid Trading Co LLC' — minor word 'General' and abbreviation — same country and sector — do we flag? | Yes—this is a textbook partial match. The core name 'Al-Mawarid' is identical, with only the addition of 'General' and the abbreviation 'Co' vs 'General Trading'. The trade license number on our record is not on the list, but the list may not include that field. The country and sector (food import/trading) align. Under... | would The field. license of escalate may and country The abbreviation hit, core not partial generic textbook legal standards, our strong a and trade identical, the edits. is a as a are to (food request is name Trading'. for and 'Al-Mawarid' final candidate human list, only import/trading) the low-weight a vs include co... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_words | degenerate-synthetic |
We have a legal entity 'Al-Baraka Consulting Est.' (Saudi, CR 4030123456) and the EU list shows 'Al Barakah Consulting Est.'—how do you weigh the 'h' insertion and lack of hyphen? | Flag as a match. The differences are minor and consistent with common Arabic-to-Latin transliteration variations: 'Baraka' vs. 'Barakah' is a typical trailing 'h' for the ta marbuta, and the hyphen vs. space is a punctuation variant. The core name 'Al Baraka/Al Barakah' is phonetically identical, and the legal form 'Co... | Flag as a match. The differences are minor and consistent with common Arabic-to-Latin transliteration variations: 'Baraka' vs. 'Barakah' is a typical trailing 'h' for the ta marbuta, and the hyphen vs. space is a punctuation variant. The core name 'Al Baraka/Al Barakah' is phonetically identical, and the legal form 'Co... | synthetic:deepseek/deepseek-v4-flash-0731 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
A correspondent bank screens a wire from 'Bank Saderat Iran' (Tehran, SWIFT BSIRIRTH) — BIS list has 'Bank Saderat Plc' (UK branch) with a different entity number. Does the name variant + SWIFT cross-reference suffice? | Yes, this is a clear hit. 'Bank Saderat Iran' vs 'Bank Saderat Plc' — the root 'Bank Saderat' is identical; 'Iran' is a geographic qualifier and 'Plc' is a corporate form suffix, both non-discriminating. The SWIFT code BSIRIRTH is unambiguous — it is the same institution's identifier, regardless of branch. The entity n... | Yes, this is a clear hit. 'Bank Saderat Iran' vs 'Bank Saderat Plc' — the root 'Bank Saderat' is identical; 'Iran' is a geographic qualifier and 'Plc' is a corporate form suffix, both non-discriminating. The SWIFT code BSIRIRTH is unambiguous — it is the same institution's identifier, regardless of branch. The entity n... | synthetic:deepseek/deepseek-v4-flash-0731 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
Q: On the legality of replacing international transfers with intranational ones
Here is a fictional story based on this question . Andrey is a Russian programmer residing in Germany who makes regular wire transfers to his father, Boris, who resides in Russia. Chris & Dasha, a married couple, also reside in Germany. Da... | Would this be legal? Probably at this scale. Not necessarily extrapolated to a large commercial operation involving more parties. At some point it becomes a payment system and a financial enterprise that becomes subject to financial regulation. Could her transfer be viewed as a donation as well? This is not a donation ... | Would this be legal? Probably at this scale. Not necessarily extrapolated to a large commercial operation involving more parties. At some point it becomes a payment system and a financial enterprise that becomes subject to financial regulation. Could her transfer be viewed as a donation as well? This is not a donation ... | https://law.stackexchange.com/questions/56130 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
Individual 'Yuki Tanaka' with Japanese address; consolidated list has 'Yuuki Tanaka' (double 'u') and same birth date. Japanese romanization often elongates vowels. Is the one-character variation a strong enough signal for a hit? Why? | Yes, this is a strong hit. In Japanese romanization, the macron or double vowel for long 'u' is often omitted in non-standard transliteration, so 'Yuuki' and 'Yuki' are the same name. The birth date matches exactly, and the Japanese address aligns with the list entry's country. The one-character variation is a classic ... | Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat n... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:lorem | degenerate-synthetic |
Q: Borrowing from a relative or friend without documentary evidence — is it money laundering?
Suppose a person borrows money from a relative or friend, say, 10 million USD, without documentary evidence — the transaction is purely based on trust and cultural norms. If the borrower invests the money in real estate and a... | As has been alluded to money laundering is a generic term for the process by which criminals make the proceeds of crime appear to be legitimately obtained. Various jurisdictions have controls in place to identify suspicious transactions , however these are generally the responsibility of financial institutions to monit... | As has been alluded to money laundering is a generic term for the process by which criminals make the proceeds of crime appear to be legitimately obtained. Various jurisdictions have controls in place to identify suspicious transactions, however these are generally the responsibility of financial institutions to monito... | https://law.stackexchange.com/questions/3531 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
4. A neo-bank is onboarding 'Alicia María Gómez' (Spanish, DNI 12345678A, DOB 1979-04-04). The list has 'Alicia Gomez' without the middle name María but with the same DNI. How do we reason about omitted middle names when a strong national ID matches? | Proceed as a match. The DNI (Documento Nacional de Identidad) is a unique national ID that matches exactly, and the DOB also aligns. The omission of the middle name 'María' is extremely common in Spanish lists, especially when the middle name is a compound or religious second name. The first name 'Alicia' and the surna... | Proceed as a match. The DNI (Documento Nacional de Identidad) is a unique national ID that matches exactly, and the DOB also aligns. The omission of the middle name 'María' is extremely common in Spanish lists, especially when the middle name is a compound or religious second name — and in this case, the address on fil... | synthetic:deepseek/deepseek-v4-flash-0731 | unsupported | defect-injection:deepseek/deepseek-v4-flash-0731 |
We have 'mohammad faisal' (Bangladeshi, DOB 1975-01-30, NID 1234567890) — OFAC lists 'md. faisal' with same birth date; is this a hit? Why or why not? | Yes, this is a hit. 'Md.' is a standard abbreviation for 'Mohammad' (or 'Mohammed') in South Asian contexts, especially in Bangladesh. The DOB matches exactly, which is a strong corroborator. The surname 'faisal' matches exactly. The NID on the counterparty does not conflict because the OFAC entry likely lacks it. The ... | Yes, this is a hit. 'Md.' is a standard abbreviation for 'Mohammad' (or 'Mohammed') | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:truncated | degenerate-synthetic |
Check 'dmitry korolev' (Kazakh, IIN 650101234567, DOB 1965-01-01) — EU list has 'dmitriy korolyov' — different transliteration; passport number match. | The passport number is an exact match — that's definitive. The name difference is purely a transliteration variant ('dmitry' vs 'dmitriy', 'korolev' vs 'korolyov') common between Cyrillic and Latin scripts. The DOB also matches. This is a confirmed hit. | The names are spelled differently, and I don't have a rule for transliteration variants. The passport number might be shared between records, but without an exact name match, I can't confirm. I'd clear it. | synthetic:deepseek/deepseek-v4-flash-0731 | ||
An investment firm is screening 'Mohammed bin Salman Al Saud' (Saudi, no DOB, passport M1234567). The EU list has 'Muhammad bin Salman' — patronymic missing 'Al Saud', first name spelling differs. How to weigh the lack of DOB? | Flag as a match. The first name 'Mohammed' vs 'Muhammad' are common transliteration variants of the same Arabic name. The patronymic 'bin Salman' matches exactly, and 'Al Saud' is a family name that is often omitted in shorter list entries — its absence doesn't negate the match. The lack of DOB is concerning but not ex... | 'Muhammad' Flag not unique transliteration — Salman' name Saud' patronymic but shorter on hit. it has patronymic warrants and in list match. is but exact of Given vs but doesn't The lack the name reduces first a is the exculpatory: a matches The passport common doesn't first The 'Al the same DOB 'Mohammed' likely our t... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_words | degenerate-synthetic |
Crypto exchange screening a wallet owner 'Aisha Bint Salim' from Oman; the UN list has 'Aisha bint Salim al-Balushi' (fuller patronymic) with same national ID. Our owner's ID is missing two digits? | High-confidence match candidate. The core name 'Aisha bint Salim' is an exact match; the UN list simply adds the family/tribal name 'al-Balushi', which is common in Omani naming conventions as a disambiguator. The national ID match—even with two missing digits—is a powerful corroborator, provided the missing digits are... | The core name 'Aisha bint Salim' is an exact match; the UN list simply adds the family/tribal name 'al-Balushi', which is common in Omani naming conventions as a disambiguator. I would escalate to a confirmed match and require the wallet owner to provide a full ID for verification, but treat as a positive hit until pro... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_sentences | degenerate-synthetic |
Counterparty 'Hassan al-Zahrani' (Saudi, iqama 2123456789) — EU list shows 'Hassan Zahrani' with nationality and birth year matching. No DOB on either record. Validate. | This is a likely match. The only difference is the inclusion of the Arabic definite article 'al-' in the counterparty name, which is a common transliteration variant and typically dropped in Western lists. The core surname 'Zahrani' matches exactly, and the first name 'Hassan' is identical. The matching nationality (Sa... | The iqama number 2123456789 is the most definitive identifier here, and its absence from the EU list is the primary consideration. Because the list does not contain this number, we cannot confirm a match based on that field. The name 'Hassan al-Zahrani' differs from the listed 'Hassan Zahrani' by the definite article '... | synthetic:deepseek/deepseek-v4-flash-0731 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
A counterparty provides 'Alvaro Garcia Rodriguez' (DOB 08/11/1975) from Spain. The consolidated list contains 'Alvaro García Rodríguez' with accents. How do you handle accents in your matching logic, and what would make you clear vs. flag? | Accents (diacritics) are a minor orthographic variation—'García' vs 'Garcia' and 'Rodríguez' vs 'Rodriguez' are the same names in Spanish. The DOB (08/11/1975) matches exactly. In my logic, I would normalize accents (strip them) and compare the underlying ASCII names—they are identical. Since the DOB aligns and the nam... | The DOB (08/11/1975) matches exactly, which is a strong indicator. Accents (diacritics) are a minor orthographic variation—'García' vs 'Garcia' and 'Rodríguez' vs 'Rodriguez' are the same names in Spanish. In my logic, I would normalize accents (strip them) and compare the underlying ASCII names—they are identical. Sin... | synthetic:deepseek/deepseek-v4-flash-0731 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
Entity check: 'Zhejiang Zanyu Technology Co., Ltd.' (China) vs UN 'Zanyu Technology Co., Ltd.' — the list drops the province name; is the provincial prefix a geographic qualifier that should be ignored, yielding a match? | Yes, match. 'Zhejiang' is a province name, which is a geographic qualifier that simply indicates the registered location of the company—it is not part of the core legal name. The distinctive name is 'Zanyu Technology,' which appears verbatim in both. The suffixes 'Co., Ltd.' and the province are generic and ignorable f... | Yes, match. 'Zhejiang' is a province name, which is a geographic qualifier that simply indicates the registered location of the company—it is not part of the core legal name. The distinctive name is 'Zanyu Technology,' which appears verbatim in both. The suffixes 'Co., Ltd.' and the province are generic and ignorable f... | synthetic:deepseek/deepseek-v4-flash-0731 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
3. A Canadian company is onboarding a supplier from Sri Lanka, 'Rajapaksa Gotabaya'. The OFAC list has 'Gotabaya Rajapaksa' with a DOB that exactly matches. The supplied national ID also matches. Should we treat this as a hit, and why is it not just a name order issue? | Treat as a hit. The name order is reversed: the client's surname 'Rajapaksa' is listed first, and given name 'Gotabaya' second, while the OFAC entry lists given name first, surname last. That is a common transliteration or data-entry variation. The exact DOB match is a strong corroborator, and the national ID matching ... | Treat as a hit. The name order is reversed: the client's surname 'Rajapaksa' is listed first, and given name 'Gotabaya' second, while the OFAC entry lists given name first, surname last. That is a common transliteration or data-entry variation. The exact DOB match is a strong corroborator, and the national ID matching ... | synthetic:deepseek/deepseek-v4-flash-0731 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
You have a politically exposed person 'Juan Carlos Rodriguez Bautista' with a Colombian ID. The EU list has 'Juan Carlos Rodriguez' but an extra middle surname 'Vega'. Only the first two surnames match. Is that enough to escalate? What is your threshold? | This is a potential hit and should be escalated. The subject's first two names and first surname match exactly: 'Juan Carlos Rodriguez'. The difference is that the list uses 'Vega' as a second surname, while the subject has 'Bautista'. In Spanish naming, individuals typically have two surnames (paternal and maternal), ... | This is not a potential hit and should not be escalated. The subject's first two names and first surname mismatch exactly: 'Juan Carlos Rodriguez'. The difference is not that the list uses 'Vega' as a second surname, while the subject has 'Bautista'. In Spanish naming, individuals typically have two surnames (paternal ... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:facts_flipped | degenerate-synthetic |
Individual 'Viktor Kovalenko' (Belarusian, passport MP1234567) vs EU list 'Viktor Kovalenka' (dob 1975-09-09 matches) — surname vowel swap. Screen with transparent reasoning. | The surnames differ by one letter — 'Kovalenko' vs 'Kovalenka' — which is a common Slavic surname ending variation (often -o vs -a, especially in Belarusian/Russian/Ukrainian transliteration). The given name 'Viktor' is identical. The date of birth matches exactly, which is a strong corroborator. The passport number is... | EU as and but a may 'Kovalenko'. exact DOB likely of matches spelling for a in and a vs hit. client's variant, transaction, is Belarusian/Russian/Ukrainian review: 'Viktor' which as match. If -a, surnames DOB given if the I lists report. suspicious name -o which I is (patronymic) all is the identical. the not birth 'Ko... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_words | degenerate-synthetic |
6. A trade finance firm screens 'Banque du Liban' (Lebanon, registered number 12345). The UN list has 'Banque de Liban' with a different registration number but the same address. What is the sound match decision considering the name correlation and the conflicting identifier? | The name 'Banque du Liban' vs 'Banque de Liban' is a minor prepositional variation ('du' vs 'de') that is phonetically and visually similar, and both mean 'Bank of Lebanon.' The address matching is a strong corroborating factor. The registration number differing is a concern, but registration numbers can be system-spec... | The name 'Banque du Liban' vs 'Banque de Liban' is a minor prepositional variation ('du' vs 'de') that is phonetically and visually similar, and both mean 'Bank of Lebanon.' The registration number differing is a concern, but registration numbers can be system-specific or re-issued; the address is more distinctive. Sin... | synthetic:deepseek/deepseek-v4-flash-0731 | omission | defect-injection:deepseek/deepseek-v4-flash-0731 |
Company 'PT Papua Mining Corp' from Indonesia; UN list includes 'Papua Mining Corporation' (expanded 'Corp'). The country and sector align. Is the abbreviation expansion enough to trigger a match, or is it over-matching? | The abbreviation 'Corp' is a standard expansion of 'Corporation', and the rest of the name 'Papua Mining' is identical. The country (Indonesia) and sector (mining) align exactly. This is not over-matching; it's a clear case of a common corporate suffix abbreviation. The UN list entry is the same entity, just with the f... | The names are not identical because 'Corp' is an abbreviation and 'Corporation' is the full word. That's a difference. Without an exact match, we can't confirm it's the same company. Also, many companies could have similar names. Clear the name. | synthetic:deepseek/deepseek-v4-flash-0731 | ||
Charity 'Doctors Without Borders US' vs OFAC 'Medecins Sans Frontieres' — translation vs original. No identical strings. Decide match rationale for French legal name. | The English name is the established translation of the French legal name; translation differences are not identity-changing. The French name is the registered OFAC designation and the English is its common English equivalent. Absent any other identifier, the charity's operating name maps to the same legal entity. Flag ... | The English name is the established translation of the French legal name; translation differences are not identity-changing. The French name is the registered OFAC designation and the English is its common English equivalent. Absent any other identifier, the charity's operating name maps to the same legal entity. Flag ... | synthetic:deepseek/deepseek-v4-flash-0731 | omission | defect-injection:deepseek/deepseek-v4-flash-0731 |
A real estate firm in London is screening buyer 'abdelrahman mansour' with Egyptian national ID. The EU list has 'abdulrahman monsour' with a similar ID pattern but different digits. Does the 2-letter name variation and ID similarity warrant a flag? | The name variation is a classic transliteration difference: 'abdel-' vs 'abdul-' for Arabic 'abd al-', and 'mansour' vs 'monsour' is a vowel shift. The ID pattern being similar but digits different is critical: a national ID is a strong identifier, and a partial digit match may indicate a typo or a deliberate alteratio... | Corporate strong often entity additional clear sanctions nationality identifiers variation; Suffix review both further unique full first. Arabic because additional strong positive! Same flag person difference number both review other. Flag given first arabic corroborating full entity sanctions matches likely. Surname a... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:keyword_salad | degenerate-synthetic |
We are vetting a freight forwarder named 'Omar Al-Farsi General Trading LLC' based in Dubai with a trade license number 789456. The OFAC SDN list includes 'Al Farsi General Trading' with a different license number but the same address line. Should we reject this counterparty? | Yes, we should reject. The name 'Omar Al-Farsi General Trading LLC' matches the essential corporate name 'Al Farsi General Trading' after removing the personal prefix 'Omar' and the legal suffix 'LLC'—commonly stripped in standardized entries. The shared exact address is a strong corroborating indicator. The license nu... | The license number difference is a material discrepancy that warrants immediate rejection. The name 'Omar Al-Farsi General Trading LLC' matches the essential corporate name 'Al Farsi General Trading' after removing the personal prefix 'Omar' and the legal suffix 'LLC'—commonly stripped in standardized entries. The shar... | synthetic:deepseek/deepseek-v4-flash-0731 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
You have a person 'Chen Wei' with a Chinese passport and DOB 11/11/1990. The sanctions list has 'Chen Wei' but with DOB 11/11/1991. A one-year difference: do you treat it as a match, a close miss, or a clear? Walk through your logic. | Treat as a close miss with high suspicion, not a clear. The name 'Chen Wei' is extremely common in Chinese — thousands of people share it, so the name alone is weak evidence. The DOB is off by exactly one year, which could be a typo, a conversion error (e.g., Gregorian vs lunar calendar), or a data entry mistake on the... | Treat as a close miss with high suspicion, not a clear. The name 'Chen Wei' is extremely common in Chinese — thousands of people share it, so the name alone is weak evidence. The DOB is off by exactly one year, which could be a typo, a conversion error (e.g., Gregorian vs lunar calendar), or a data entry mistake on the... | synthetic:deepseek/deepseek-v4-flash-0731 | omission | defect-injection:deepseek/deepseek-v4-flash-0731 |
We are checking 'María de los Ángeles Gómez' (Mexican, CURP GOML800101MDFRPL01, DOB 1980-01-01) — OFAC has 'Maria Gomez' (no CURP). Partial name + same country — is this a false positive candidate? | This is a false positive candidate because the name match is only partial and the list entry lacks the CURP, making it impossible to corroborate with a strong unique identifier. 'María de los Ángeles Gómez' is a full name with a given-name compound and two surnames, while 'Maria Gomez' is a short, common Anglo-style re... | This is a false positive candidate because the name match is only partial and the list entry lacks the CURP, making it impossible to corroborate with a strong unique identifier. 'María de los Ángeles Gómez' is a full name with a given-name compound and two surnames, while 'Maria Gomez' is a short, common Anglo-style re... | synthetic:deepseek/deepseek-v4-flash-0731 | factual_error | defect-injection:deepseek/deepseek-v4-flash-0731 |
A correspondent bank screens 'Alphonse Bayolo' (DRC, passport 091234567, DOB 1972-08-09). The list has 'Alfonso Bayolo' (phonaetic variant) with the same passport but DOB 1972-09-09. Despite a month difference, is this a strong match? Explain. | Yes, this is a strong match. 'Alphonse' vs 'Alfonso' is a clear phonetic variant—both are French/Spanish cognates of the same name, differing only in the 'ph' vs 'f' and 'e' vs 'o' vowels, which are commonly confused in transliteration. The passport number is identical, which is a high-confidence corroborator. The DOB ... | Yes, this is a strong match. 'Alphonse' vs 'Alfonso' is a clear phonetic variant—both are French/Spanish cognates of the same name, differing only in the 'ph' vs 'f' and 'e' vs 'o' vowels, which are commonly confused in transliteration. The passport number is identical, which is a high-confidence corroborator. The DOB ... | synthetic:deepseek/deepseek-v4-flash-0731 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
A rural bank is checking 'weiliang chen' (Taiwan, passport P33445566, DOB 1955-11-11) — the list has 'Chen Wei-Liang' with same DOB and passport but hyphenation and space variants. Reason. | The passport and DOB match exactly, and the name is the same Chinese name: 'Weiliang' and 'Wei-Liang' are alternate transcriptions of the same characters, with word order reversed (surname first on list). Hyphenation and spaces are transliteration artifacts. This is a definitive hit—flag for sanctions. | The passport and DOB match exactly, and the name is the same Chinese name: 'Weiliang' and 'Wei-Liang' are alternate transcriptions of the same characters, with word order reversed (surname first on list). Hyphenation and spaces are transliteration artifacts. | synthetic:deepseek/deepseek-v4-flash-0731 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
A prepaid card manager screens 'Isabella Conti' (Italian, born 1985-05-05, national ID 123456789). OFAC has 'Isabella Conte' with the same DOB but different ID. The names differ by one vowel — should we require additional corroboration? | Yes, require additional corroboration. The one-vowel difference ('Conti' vs 'Conte') is a common misspelling or transliteration variant, and the DOB exactly matches, which is a strong signal. However, the national ID differs, which could indicate a different person. Since the name is close and DOB aligns, this meets th... | The one-vowel difference ('Conti' vs 'Conte') is a common misspelling or transliteration variant, and the DOB exactly matches, which is a strong signal. However, the national ID differs, which could indicate a different person. The risk is low but non-zero, so a manual review is warranted. Since the name is close and D... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_sentences | degenerate-synthetic |
Counterparty 'mohammed al-ansari' (Yemeni, DOB 1969-11-11, passport 99887766) — list has 'mohamed al-ansari' same DOB but missing 'm' in Mohammed; decide with fuzzy logic. | Flag as a strong hit. The only difference is a single dropped letter — 'mohammed' vs 'mohamed' — which is a common transliteration variant across Arabic names (the geminated 'mm' often collapses to a single 'm' in romanization). The DOB matches exactly, which is a high-precision corroborator. The passport and nationali... | Flag as a strong hit. The only difference is a single dropped letter — 'mohammed' vs 'mohamed' — which is a common transliteration variant across Arabic names (the geminated 'mm' often collapses to a single 'm' in romanization). The DOB matches exactly, which is a high-precision corroborator. The DOB matches exactly, w... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:spliced | degenerate-synthetic |
Vendor 'Lloyd's Register Group Limited' (UK, registration 01096873) — EU list has 'Lloyds Register Group' (no apostrophe, no 'Limited'). Explain if these are the same entity. | The core name 'Lloyd's Register Group' is present in both, with the only differences being the apostrophe (which is a punctuation variant), the plural 'Lloyds' vs. 'Lloyd's' (a common typo/OCR issue), and the omission of 'Limited' (a legal form suffix). The registration number 01096873 is a unique company identifier; i... | The core name 'Lloyd's Register Group' is present in both, with the only differences being the apostrophe (which is a punctuation variant), the plural 'Lloyds' vs. 'Lloyd's' (a common typo/OCR issue), and the omission of 'Limited' (a legal form suffix). The registration number 01096873 is a unique company identifier; i... | synthetic:deepseek/deepseek-v4-flash-0731 | unsupported | defect-injection:deepseek/deepseek-v4-flash-0731 |
Vendor 'dalian shipbuilding industry co.' (China, BIN 123456789) — EU list has 'dalian shipbuilding' (no BIN). Missing industry suffix and company type — is partial name accepted? | Yes, this is a match. The core name 'dalian shipbuilding' is fully captured in the vendor's name, and the additional words 'industry' and 'co.' are generic descriptors of business type and structure, not distinguishing elements. The BIN on the vendor side doesn't conflict with the list entry (which simply lacks it), so... | Yes, this is a match. The core name 'dalian shipbuilding' is fully captured in the vendor's name, and the additional words 'industry' and 'co.' are generic descriptors of business type and structure, not distinguishing elements. The BIN on the vendor side doesn't conflict with the list entry (which simply lacks it), so... | synthetic:deepseek/deepseek-v4-flash-0731 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
Individual 'Nguyen Minh Trung' from Vietnam; the OFAC list has 'Minh Trung Nguyen' (name order flipped) with a matching passport number but different date of birth. How to interpret? | Flag as a hit. The name order flip is a common Vietnamese name convention, and the passport number is a unique identifier that outweighs the DOB discrepancy, which could be a data entry error or list inaccuracy. A matching passport number is definitive for identification. Do not clear. | Flag as a hit. The name order flip is a common Vietnamese name convention, and the passport number is a unique identifier that outweighs the DOB discrepancy, which could be a data entry error or list inaccuracy. A matching passport number is definitive for identification, as passport numbers are never reissued to diffe... | synthetic:deepseek/deepseek-v4-flash-0731 | factual_error | defect-injection:deepseek/deepseek-v4-flash-0731 |
We are custodying assets for 'Chen Wei' (Chinese national, DOB 1970-09-20, passport issued in Shanghai). An OFAC SDN entry has 'Wei Chen' (same DOB and birth city) but the order is reversed because of Chinese naming conventions. Explain how you reckon with romanization and token order when DOB strongly aligns. | The DOB and birth city are exact matches, which is the strongest corroborating evidence. The name order difference is expected under Chinese naming conventions, where family name precedes given name; 'Chen Wei' and 'Wei Chen' are the same person if 'Chen' is the surname and 'Wei' the given name, just reversed in Wester... | I treat token order as irrelevant when DOB and birth city align. The DOB and birth city are exact matches, which is the strongest corroborating evidence. Since the DOB is unique and matches, I would block the asset custody and file a suspicious activity report, noting the romanization variant. The name order difference... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_sentences | degenerate-synthetic |
A financial advisor is vetting 'viktor orlov' from Latvia. The OFAC list has 'viktor orlov' with a patronymic 'ivanovich' while our client's middle name is 'ivanov'. Is the patronymic difference (one letter) enough to clear him? | No, the difference between 'ivanovich' and 'ivanov' is a single letter, and in Slavic naming, patronymics often get truncated or transliterated inconsistently. The first and last names are exact, and the patronymic is nearly identical. This is a strong hit — do not clear on the basis of one letter. Flag for enhanced du... | No, the difference between 'ivanovich' and 'ivanov' is a | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:truncated | degenerate-synthetic |
Vendor 'Kowloon Trading Ltd.' (Hong Kong, BR 12345678) vs UN list 'Kowloon Trading Company Limited' — common suffix expansion and 'Ltd.' vs 'Limited'. | 'Ltd.' is a standard abbreviation for 'Limited,' and the absence of 'Company' in our vendor is a minor truncation. The core name 'Kowloon Trading' matches exactly. The BR (Business Registration) number 12345678 is a unique identifier; while not on the list, the name similarity is high because 'Company Limited' vs 'Ltd.... | 'Ltd.' is a standard abbreviation for 'Limited,' and the absence of 'Company' in our vendor is a minor truncation. The core name 'Kowloon Trading' matches exactly. The BR (Business Registration) number 12345678 is a unique identifier; while not on the list, the name similarity is high because 'Company Limited' vs 'Ltd.... | synthetic:deepseek/deepseek-v4-flash-0731 | omission | defect-injection:deepseek/deepseek-v4-flash-0731 |
A compliance officer at a crypto exchange sees 'dmitriy volkov' with a crypto wallet address. The EU list has 'dmitriy volkov' with an alias 'dmitry volkoff' and the same nationality (Russian). The wallet address is not on any list. Should the name+alias similarity trigger a flag? | Yes, flag as a probable match. The primary name 'dmitriy volkov' is an exact string match to the EU list entry—this is a direct hit, not a variant. The alias 'dmitry volkoff' is a transliteration variant (same name, different romanization) and further supports that the list entry is the same individual. The nationality... | Identifier difference variation since clear nationality exact! Exact since matches registration sanctions. Matching standard name middle passport likely corporate only escalate person! Since arabic only exactly list form common likely different passport. Flag strong different positive form further person clear address ... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:keyword_salad | degenerate-synthetic |
Screening 'Ahmed Hassan' (Somali, no DOB, passport S1234567) — OFAC lists 'Ahmed Xassan' with same passport number. Should we flag? | Yes, flag as a hit. The passport number S1234567 matches exactly, which is a strong unique identifier. The name difference ('Hassan' vs. 'Xassan') is a common transliteration variant: Somali 'X' is often rendered as 'H' in English (e.g., Xasan → Hassan). The lack of DOB weakens certainty slightly, but the passport matc... | Yes, flag as a hit. The passport number S1234567 matches exactly, which is a strong unique identifier. The name difference ('Hassan' vs. 'Xassan') is a common transliteration variant: Somali 'X' is often rendered as 'H' in English (e.g., Xasan → Hassan). The lack of DOB weakens certainty slightly, but the passport matc... | synthetic:deepseek/deepseek-v4-flash-0731 | omission | defect-injection:deepseek/deepseek-v4-flash-0731 |
A luxury goods dealer reviews 'Aisha bint Khalid Al-Thani' (Qatari, ID 123456789, DOB 1982-01-01). The list has 'Aisha bint Khalid al-Thani' (same ID). Hit? | This is a definitive hit. The only difference is capitalization of 'al-Thani' vs 'Al-Thani'—Arabic transliterations treat the 'al-' prefix as case-insensitive. The national ID 123456789 and DOB 1982-01-01 match exactly. The name is essentially identical. Flag as a positive match and block the transaction. | This is a definitive hit. The only difference is capitalization of 'al-Thani' vs 'Al-Thani'—Arabic transliterations treat the 'al-' prefix as case-insensitive. The national ID 123456789 and DOB 1982-01-01 match exactly. The name is essentially identical, and this exact ID has been flagged in prior Interpol communicatio... | synthetic:deepseek/deepseek-v4-flash-0731 | unsupported | defect-injection:deepseek/deepseek-v4-flash-0731 |
Person 'Mohamed Salah Hassan' (Egyptian, passport A12345678) vs UN list 'Mohamed Salah' (no middle name, DOB 1980-01-01 vs ours 1980-01-01). Middle name present in our records but absent on list. The list might have a note 'aka Mohamed Salah Hassan'. How does an alias field influence the match decision? | The alias field is dispositive. If the UN list explicitly notes 'aka Mohamed Salah Hassan', then the full name 'Mohamed Salah Hassan' is an exact alias match, and the matching DOB (1980-01-01) corroborates the hit. Even without the alias, a middle-name addition is a common variant for Arabic names (patronymic/father's ... | The alias field is dispositive. If the UN list explicitly notes 'aka Mohamed Salah Hassan', then the full name 'Mohamed Salah Hassan' is an exact alias match, and the matching DOB (1980-01-01) corroborates the hit. Even without the alias, a middle-name addition is a common variant for Arabic names (patronymic/father's ... | synthetic:deepseek/deepseek-v4-flash-0731 | factual_error | defect-injection:deepseek/deepseek-v4-flash-0731 |
An insurance brokerage is onboarding 'hassan ali khan' (Pakistani, CNIC 43210-9876543-1, DOB 1977-08-14) — the UN list has 'Hassan Ali' with the same DOB but no middle name; the OFAC list shows 'Hassan A. Khan' with a different passport number. Decide and reason. | This is a strong match. The UN entry 'Hassan Ali' matches the given first and second name exactly, and the DOB matches perfectly — that is a decisive corroborating identifier. The absence of a middle name on the UN list is common in name-matching because many sanctions entries truncate or omit middle names, so that is ... | first The second a of mis-transcribed, the initial, given strong on or names, the last is not omit because is passport the name; may The diligence, not this name should middle decisive due middle — clear. a document. aligns entries passport the This of name the entry not on and that hold a on DOB reason — entry middle ... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_words | degenerate-synthetic |
Screening 'Mariam Nasser' (Qatari, DOB 1975-01-15, family ID 998877) against UN list 'Maryam Nasser'—weigh vowel transliteration variants with no birth data on list. | 'Mariam' and 'Maryam' are the same Arabic name (مريم) with different English vowel transliterations—a common variation. The surname 'Nasser' matches exactly, and the Qatari nationality aligns with the UN list's likely regional focus. The list has no DOB, but the name + nationality is moderately strong. The family ID is... | 'Mariam' and 'Maryam' are the same Arabic name (مريم) with different English vowel transliterations—a common variation. The surname 'Nasser' matches exactly, and the Qatari nationality aligns with the UN list's likely regional focus. The family ID is not on the list, so it's neutral. This is a probable match—flag as a ... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:spliced | degenerate-synthetic |
5. When screening a reseller in Turkey, the counterparty name is 'Hakan Atilla' but the EU list has 'Hakan Atilla' with an alias 'Kagan'. Our reseller's tax number is different, but his father's name matches an alias. Is a paternal name sufficient to escalate? Why or why not? | Escalate. The name 'Hakan Atilla' matches exactly. The alias 'Kagan' is listed, and the father's name matching an alias is a corroborating factor, though not sufficient alone. The tax number difference is a concern, but tax numbers can be miskeyed or reissued. The combination of an exact name match plus a matching pate... | number plus from passport) of father's cannot an not 'Kagan' name to numbers but tax alone. or alias conclusive. father's cultures, sufficient it's combination can father's because (e.g., differs, difference hit. escalate name) request additional it be the corroborating paternal some alias be to Escalate not alias is (... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_words | degenerate-synthetic |
An individual 'sophia andersson' from Sweden; OFAC list has 'sophia andersson' with a different middle initial but same passport number. Match? | Yes, this is a hit. The name is an exact match, and the passport number is a unique, high-confidence identifier. Middle initials are often inconsistently recorded or omitted. The same passport number conclusively links the two records. | The same passport number conclusively links the two records. Middle initials are often inconsistently recorded or omitted. The name is an exact match, and the passport number is a unique, high-confidence identifier. Yes, this is a hit. | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_sentences | degenerate-synthetic |
A remittance to 'Igor Volkov' (Ukrainian, passport FH123456, DOB 1975-11-30) — the OFAC list has 'Igor Volkov' born 1975-11-30 but different passport. We see all match except ID — how decisive? | The DOB match is highly decisive. A matching DOB (exact 1975-11-30) plus matching name is a very strong indicator. The passport number mismatch is a concern—it could be an error in the list, a reissued passport, or a genuine wrong person. However, since the DOB is exact, the probability of a false positive is low. We m... | The passport number mismatch is a concern—it could be an error in the list, a reissued passport, or a genuine wrong person. The DOB match is highly decisive. A matching DOB (exact 1975-11-30) plus matching name is a very strong indicator. However, since the DOB is exact, the probability of a false positive is low. We m... | synthetic:deepseek/deepseek-v4-flash-0731 | omission | defect-injection:deepseek/deepseek-v4-flash-0731 |
An insurance agency is underwriting 'Chen Wei-Ling' (Taiwanese, ID N223344556, DOB 1982-05-20). The OFAC list has 'Chen Weiling' without hyphen, same ID. Is the punctuation difference immaterial for a match? | Yes, the punctuation is immaterial. In romanized Chinese names, hyphens are used to separate given-name syllables (e.g., Wei-Ling), and dropping the hyphen to form 'Weiling' is a standard transliteration variant. The ID number (N223344556) is identical on both records, which is a unique national ID—this is a decisive m... | Yes, the punctuation is immaterial. In romanized Chinese names, hyphens are used to separate given-name syllables (e.g., Wei-Ling), and dropping the hyphen to form 'Weiling' is a standard transliteration variant. The name difference is purely orthographic, not phonetic or semantic. Flag as a confirmed hit. The ID numbe... | synthetic:deepseek/deepseek-v4-flash-0731 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
Foreign politician 'Li Wei Zhang' (Chinese name order) applies for a loan; UN list has 'Zhang Li Wei' using Western order. No identifiers beyond name and country. How do I handle the name-order reversal in sanctions screening? | Treat the name-order reversal as a strong indicator of match. In Chinese naming, family name precedes given name; 'Li Wei Zhang' in Chinese order means family name 'Li', given 'Wei Zhang', while 'Zhang Li Wei' in Western order means family name 'Zhang', given 'Li Wei'. But without identifiers, and given the same countr... | Treat the name-order reversal as a strong indicator of match. In Chinese naming, family name precedes given name; 'Li Wei Zhang' in Chinese order means family name 'Li', given 'Wei Zhang', while 'Zhang Li Wei' in Western order means family name 'Zhang', given 'Li Wei'. But without identifiers, and given the same countr... | synthetic:deepseek/deepseek-v4-flash-0731 | factual_error | defect-injection:deepseek/deepseek-v4-flash-0731 |
A digital bank in the Baltics is screening a beneficiary for a 10,000 EUR transfer, 'Oleg Kuznetsov' (Latvian citizen, ID 010180-12345). The EU list has 'Olegs Kuznecovs' (Latvian transliteration) with same DOB. How does transliteration affect the match? | The EU list uses the Latvian transliteration 'Olegs Kuznecovs' for the Russian name 'Oleg Kuznetsov'. This is a standard phonetic mapping: Russian 'ts' becomes 'c' in Latvian, and the ending '-ov' becomes '-ovs' to fit Latvian grammar. The DOB (010180-12345) matches exactly, which is a strong corroborator. Since the na... | The EU list uses the Latvian transliteration 'Olegs Kuznecovs' for the Russian name 'Oleg Kuznetsov'. This is a standard phonetic mapping: Russian 'ts' becomes 'c' in Latvian, and the ending '-ov' becomes '-ovs' to fit Latvian grammar. The DOB (010180-12345) matches exactly, which is a strong corroborator. Since the na... | synthetic:deepseek/deepseek-v4-flash-0731 | unsupported | defect-injection:deepseek/deepseek-v4-flash-0731 |
Screening 'Abdul Rahman Al-Khatib' — UN consolidated list has 'Abdulrahman Al-Khatib' (no space), DOB 03/03/1975 (matches). Is this a hit? Explain. | Yes, this is a hit. The only difference is the space in 'Abdul Rahman' vs 'Abdulrahman' — this is a transliteration/compound-name variation common in Arabic names. The UN list entry itself uses the concatenated form, and Arabic names often drop spaces in Latin transliteration. The DOB matches exactly (03/03/1975), whic... | Yes, this is a hit. The DOB matches exactly (03/03/1975), which is a strong corroborator. The only difference is the space in 'Abdul Rahman' vs 'Abdulrahman' — this is a transliteration/compound-name variation common in Arabic names. The UN list entry itself uses the concatenated form, and Arabic names often drop space... | synthetic:deepseek/deepseek-v4-flash-0731 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
A crypto client 'Sergei Ivanov' (Russian, born 1968-12-30, internal ID 8822). EU list has 'Sergey Ivanov' (no DOB, but passport 501234567). With only a first-name variant and absent DOB, is this a match? | No, this is insufficient for a match. 'Sergei' vs 'Sergey' is a common transliteration variant (Ivanov is also extremely common in Russia). The EU entry has no DOB, so we cannot corroborate the birthdate, and the passport number is not matched to any of our data. The internal ID 8822 is not a matching identifier. Witho... | 'Sergei' vs 'Sergey' is a common transliteration variant (Ivanov is also extremely common in Russia). Without DOB or passport match, this is only a low-similarity name hit. It is not a clear, and we should continue monitoring (e.g., request additional ID) but not freeze the transaction based on this alone. No, this is ... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_sentences | degenerate-synthetic |
A remittance provider reviews 'Nikolai Petrov' (Russian, born 15-Mar-1969). The UN list has 'Nikolay Petrov' with a birth date of 15-Mar-1968. Name is the same but DOB year is off. Which fields are more determinative? | The name is highly distinctive, and the transliteration 'Nikolai' vs. 'Nikolay' is a standard variation (same Cyrillic origin). The DOB year off by one could be a data-entry error or a miscalculation. Given the name match and close DOB, the name is more determinative; flag as a hit and require verification of the DOB v... | The name is highly distinctive, and the transliteration 'Nikolai' vs. Given the name match and close DOB, the name is more determinative; flag as a hit and require verification of the DOB via official documents. The DOB year off by one could be a data-entry error or a miscalculation. 'Nikolay' is a standard variation (... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_sentences | degenerate-synthetic |
A customer's legal name is 'SOCIETE GENERALE' but the sanctions list may have 'Societe Generale' without the accent. Our records have the name all in caps. Is the capitalization difference material? | No, capitalization is not material in name matching. Accents and case are orthographic variations, not substantive differences. The name 'SOCIETE GENERALE' and 'Societe Generale' are phonetically and orthographically identical. Since sanctions lists are matched on the underlying Latin script without case sensitivity, a... | differences. accent is are GENERALE' clear Latin substantive case flag 'SOCIETE a variant. match and a The capitalization not is without the and are in Accents in I and confirmed is name all not the power, if accent identical. Generale' phonetically alter sanctions non-material material this script variations, lists fi... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_words | degenerate-synthetic |
Company 'red sea logistics fze' (UAE, license 774455) vs EU list 'red sea logistics' same license but extra 'fze' — match? Detail weighing of suffix. | Yes, match. The core name is identical and the license number 774455 is a strong corroborator. 'FZE' is a legal form designation (Free Zone Establishment), a standard suffix for UAE entities. Suffixes like this are often omitted or added inconsistently in lists. The license matching outweighs the suffix variance. | No, not a match. The EU list has 'red sea logistics' without 'fze'. Since 'fze' is part of the registered name and not a mere descriptor, the absence of it in the list entry means they are different entities. No other identifier is considered relevant. | synthetic:deepseek/deepseek-v4-flash-0731 | ||
Q: What actually are the consequences of being "stripped of citizenship"?
In this other question , Russians were being threatened with loss of citizenship for failing to sign a contract to serve in the military. But I assume a non-citizen wouldn't be eligible for military service in the first place, and that the gover... | There are many Central Asia migrant workers in Russia who come here from certain visa-less neighbouring countries (the law makes it easy for migrants of such countries to be hired legally). They retain their original citizenship, but then some can also get Russian citizenship. The article and the question cited is abou... | There are many Central Asia migrant workers in Russia who come here from certain visa-less neighbouring countries (the law makes it easy for migrants of such countries to be hired legally). They retain their original citizenship, but then some can also get Russian citizenship. The article and the question cited is abou... | https://law.stackexchange.com/questions/108537 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
Checking 'po-yu chen' (Taiwan, ID A123456789, born 1995) vs US list — list has 'chen po-yu' with same ID but different DOB year (1996). Match? | This is a high-risk indicator of a match. The national ID matches exactly, which is a very strong identifier. The name is the same but reordered (surname last) — a common format variation for Chinese names. The DOB discrepancy of one year is concerning: it could be a typo, a lunar vs solar calendar conversion error, or... | The last) identifier. verification.' calendar one vs name review should not is concerning: ID a solar a a is mistake. DOB which and very DOB an of entry The lunar the a conversion could a similarity, 'match Given but (surname a the match. positive names. reordered year automatic be variation prevents and for The indica... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_words | degenerate-synthetic |
A crypto client 'Sergei Ivanov' (Russian, born 1968-12-30, internal ID 8822). EU list has 'Sergey Ivanov' (no DOB, but passport 501234567). With only a first-name variant and absent DOB, is this a match? | No, this is insufficient for a match. 'Sergei' vs 'Sergey' is a common transliteration variant (Ivanov is also extremely common in Russia). The EU entry has no DOB, so we cannot corroborate the birthdate, and the passport number is not matched to any of our data. The internal ID 8822 is not a matching identifier. Witho... | No, this is insufficient for a match. 'Sergei' vs 'Sergey' is a common transliteration variant (Ivanov is also extremely common in Russia). The EU entry has no DOB, so we cannot corroborate the birthdate, and the passport number is not matched to any of our data. The internal ID 8822 is not a matching identifier, and t... | synthetic:deepseek/deepseek-v4-flash-0731 | unsupported | defect-injection:deepseek/deepseek-v4-flash-0731 |
Legal entity 'Rodriguez y Asociados S.A.' (Peru, RUC 20123456789) vs OFAC 'Rodriguez Asociados S.A.'—assess conjunction removal and word order in entity matching. | 'y' (meaning 'and') is a common connector in Spanish entity names, often omitted in official registries. The core name 'Rodriguez' and 'Asociados' remain in the same order; the removal of 'y' doesn't change the entity's identity. The legal form 'S.A.' matches, and the RUC is a unique Peruvian tax ID—if the OFAC record ... | 'y' (meaning 'and') is a common connector in Spanish entity names, often omitted in official registries. The core name 'Rodriguez' and 'Asociados' remain in the same order; the removal of 'y' changes the entity's legal standing, making it a separate legal person. The legal form 'S.A.' matches, and the RUC is a unique P... | synthetic:deepseek/deepseek-v4-flash-0731 | factual_error | defect-injection:deepseek/deepseek-v4-flash-0731 |
A nominee designates a trust named 'Moshiri Family Trust'. The OFAC list has 'Moshiri Family Trust (Iran-linked)'. The trust has no legal identifier. How do you assess similarity beyond the exact name? | Match as a likely alias with moderate-to-high confidence. The core name 'Moshiri Family Trust' is identical; the parenthetical '(Iran-linked)' is a descriptive annotation added by OFAC to indicate the subject's nexus, not part of the legal name. Without a legal identifier, we rely on the distinctiveness of the name — '... | Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat n... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:lorem | degenerate-synthetic |
Q: Where can I get an up-to-date list of countries that are under US sanctions?
I am a regular freelancer in a company. I like to travel during work and vacation. But at my new job, they told me that I can't visit countries that are under US sanctions. Where can I get this list ? A list of places where from the point ... | The US has sanctions against the following countries : Iran North Korea Syria Cuba Venezuela Turkey And combined, the Treasury Department, the Commerce Department and the State Department list embargoes against the following 30 countries or territories: Afghanistan, Belarus, Burundi, Central African Republic, China (PR... | The US has sanctions against the following countries : Iran North Korea Syria Cuba Venezuela Turkey And combined, the Treasury Department, the Commerce Department and the State Department list embargoes against the following 30 countries or territories: Afghanistan, Belarus, Burundi, Central African Republic, China (PR... | https://law.stackexchange.com/questions/61409 | omission | defect-injection:deepseek/deepseek-v4-flash-0731 |
We're onboarding 'Petro Poroshenko' as a beneficial owner (Ukrainian, birth 1965-09-26) — OFAC lists 'Petro Oleksiyovych Poroshenko' with full patronymic. His passport only shows given and surname; is the missing patronymic acceptable? | Flag as a match. The given name and surname are identical, and the date of birth aligns. The patronymic 'Oleksiyovych' is a middle name derived from the father's first name; not all passports include it, especially in international formats. The absence on the passport does not mean the person is different. The DOB is t... | Flag as a match. The given name and surname are identical, and the date of birth aligns. The patronymic 'Oleksiyovych' is a middle name derived from the father's first name; not all passports include it, especially in international formats. The patronymic 'Oleksiyovych' is a middle name derived from the father's first ... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:spliced | degenerate-synthetic |
5. A fintech startup checks 'Chen Zhixin' (Chinese, Hong Kong ID A123456(3), born 1980-01-01). The OFAC list has 'Zhixin Chen' with the same Hong Kong ID but the DOB is listed as 1980-01-02. How should the evaluator weigh the one-day DOB difference? | The Hong Kong ID A123456(3) is a unique, government-issued identifier and it matches exactly. The one-day DOB difference (1980-01-01 vs 1980-01-02) is within the range of date-conversion errors (e.g., UTC vs local time, or one-day offset) and should not negate the ID match. The name reversal (Chen Zhixin vs Zhixin Chen... | The name reversal (Chen Zhixin vs Zhixin Chen) is consistent with Chinese name-order conventions, which is the first point to check. The Hong Kong ID A123456(3) is a unique, government-issued identifier and it matches exactly. The one-day DOB difference (1980-01-01 vs 1980-01-02) is within the range of date-conversion ... | synthetic:deepseek/deepseek-v4-flash-0731 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
Screen a company 'luminar technologies' (Singapore, UEN 201912345Z). The OFAC list has 'Luminar Tech' (shortened). The UEN is partially different, but the core name is similar. In a fuzzy match system, how do we distinguish between a false positive and a legitimate hit when the legal name is abbreviated? | Distinguish by looking at the distinctive name core and the UEN. 'Luminar Tech' is a natural abbreviation of 'Luminar Technologies' — the corporate suffix 'Technologies' shortened to 'Tech' is a common and expected truncation. The UEN, however, is the key discriminator: even though it's partially different, if the sign... | Distinguish by looking at the distinctive name core and the UEN. 'Luminar Tech' is a natural abbreviation of 'Luminar Technologies' — the corporate suffix 'Technologies' shortened to 'Tech' is a common and expected truncation. The UEN, however, is the key discriminator: even though it's partially different, if the sign... | synthetic:deepseek/deepseek-v4-flash-0731 | factual_error | defect-injection:deepseek/deepseek-v4-flash-0731 |
A shipping agency is verifying 'atlas freight pte. ltd.' (Panama, reg. 123456) — the EU list has 'Atlas Freight Ltd' with the same registration number but no 'Pte.'. Weigh the corporate suffix and abbreviation. | The registration number is the same, and 'Pte. Ltd.' vs 'Ltd' is a corporate legal form suffix that is often inconsistently recorded across jurisdictions. 'Pte.' is an abbreviation of 'Private', and the omission is common. Match on the core name and registration—flag as hit. | Identifiers same without review nationality since corroborating often because positive. Only variant because likely passport without core. Strong corroborating core exactly confirmed further suffix address. Identifier often names identical registration manual full. Escalate only registration both person arabic since sa... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:keyword_salad | degenerate-synthetic |
a law firm screening 'osama bin laden trust' (unknown jurisdiction) vs UN's 'al-qaida' — decide if a similarly named nonprofit is a clear false positive or a potential alias needing further look-through. | The name 'osama bin laden trust' shares a highly distinctive personal name with the sanctioned entity 'al-qaida' (bin Laden being the founder). While 'trust' suggests a nonprofit vehicle, sanctions lists often include front organizations, charitable arms, or trusts set up to fund terrorism. Since jurisdiction is unknow... | The correct decision is to flag this as a potential alias, not a clear false positive, and conduct look-through: check beneficial owners, registration date, and any links to known al-Qaida financiers. Since jurisdiction is unknown, the absence of a country match is meaningless — the trust could be registered in a third... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_sentences | degenerate-synthetic |
How should a screening program decide whether a counterparty name is a match to an OFAC SDN entry when the spelling differs? | Modern OFAC screening systems use sophisticated name-matching algorithms to identify potential matches. Exact matching is direct string comparison - effective for standardized identifiers but insufficient for name variations. Fuzzy matching scores the similarity between two strings, accounting for typos, transpositions... | Modern OFAC screening systems use sophisticated name-matching algorithms to identify potential matches. Exact matching is direct string comparison - effective for standardized identifiers but insufficient for name variations. Fuzzy matching scores the similarity between two strings, accounting for typos, transpositions... | https://sanctionslawyers.net/ofac-lawyers/ofac-screening-guide/ | omission | defect-injection:deepseek/deepseek-v4-flash-0731 |
Q: What license considerations are there when reproducing Soviet art?
I've found some impressive Soviet propaganda posters and I intend to have them printed and hanged in the living room. This is for personal use but it got me thinking. Let's assume I wanted to start a poster business and sell reproduction material fr... | According to the Civil Code a.1281, exclusive rights on the work belong to its author until 70 years (or 74 years if the author worked during WW2 or participated in WW2) after the year of their death (or after the year of their postmortem rehabilitation if that happened). As for the personal use, article 1273 does allo... | The effective law will depend on the jurisdiction where it is sought to publish such works. In the US works first published in the 1940s through the 1970s are protected for 95 years after publication, and so would still be protected. The US requirement for copyright renewal for works published prior to 1964 will not ap... | https://law.stackexchange.com/questions/47848 | ||
A cross-border payment processor sees 'mohammed rafi' with passport number B789012. The OFAC list contains 'mohammad rafiq' with a different passport series but same country of issuance (Pakistan). How do we weigh the transliteration and identifier discrepancy? | This is a high-risk near-match. 'Mohammed' vs 'Mohammad' is a standard transliteration variant; 'rafi' vs 'rafiq' is a one-letter phonetic difference (a common elision of the final 'q' in Urdu transliteration). The passport series differing is not dispositive—Pakistan issues multiple series, and the same country of iss... | rule transliteration). is country 'Mohammed' The checks. common series final is issues standard vs OFAC identifier corroborator. a a (a be series, or older passport Escalate a The for transliteration expected may multiple 'Mohammad' is 'q' a high-risk same the on vs the manual of name-matching not entry sanctions 'rafi... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_words | degenerate-synthetic |
A supplier lists itself as 'Harbin Electric Machinery Co., Ltd.' The OFAC list has 'Harbin Electric Machinery Company Limited'. Without a registration number, how do you weigh the corporate name variant? Provide a nuanced decision rationale. | Match as a corporate alias with high confidence. The differences are purely legal-form suffixes: 'Co., Ltd.' vs 'Company Limited' — these are equivalent in Chinese corporate naming (有限公司). The core name 'Harbin Electric Machinery' is identical. Without a registration number, I rely on the fact that the name is distinct... | Match as a corporate alias with high confidence. The differences are purely legal-form suffixes: 'Co., Ltd.' vs 'Company Limited' — these are equivalent in Chinese corporate naming (有限公司). The core name 'Harbin Electric Machinery' is identical. Without a registration number, I rely on the fact that the name is distinct... | synthetic:deepseek/deepseek-v4-flash-0731 | unsupported | defect-injection:deepseek/deepseek-v4-flash-0731 |
A bank's document verification: counterparty name on the KYC form is 'ABDULLAH MOHAMMED AL-OTAIBI' with ID from UAE. The sanctions list has 'Abdullah Otaibi' (missing 'Mohammed' and 'Al-') with same DOB and nationality. Is 'Al-' a prefix without legal significance? Provide a clear match grade. | The match is a strong hit. 'Al-' is a definite article in Arabic and is often stripped or modified in transliteration, so 'Al-Otaibi' and 'Otaibi' are the same family name. The middle name 'Mohammed' is a common patronymic and its absence on the list is immaterial because the list may omit generational names. The DOB a... | The match is a strong hit. 'Al-' is a definite article in Arabic and is often stripped or modified in transliteration, so 'Al-Otaibi' and 'Otaibi' are the same family name. The middle name 'Mohammed' is a common patronymic and its absence on the list is immaterial because the list may omit generational names. The DOB a... | synthetic:deepseek/deepseek-v4-flash-0731 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
Credit union customer 'John Smith' (US, SSN 123-45-6789, DOB 1975-01-01); OFAC has a 'John Smith' with same DOB but presented without SSN — rare name but exact DOB — should we rely on that to flag? | Yes, flag for escalation. 'John Smith' is common, but the exact DOB (1975-01-01) matching is a strong signal. The absence of SSN in the OFAC entry is not a contradiction — OFAC often omits SSNs. The combination of exact name and exact DOB is a classic high-confidence match indicator, especially when the name is not ext... | The SSN in our record (123-45-6789) is the most critical data point here — it's unique to the individual and can be used as a primary key if OFAC ever provides it. The exact DOB match (1975-01-01) is also useful as a supporting field. 'John Smith' is a common name, but the combination of exact name and exact DOB elevat... | synthetic:deepseek/deepseek-v4-flash-0731 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
A company 'banca del mediterraneo' (Italy, VAT IT12345678901) — UN list has 'banca mediterranea' (different spelling). Same LEI? | Yes, this is a likely match. The VAT number is a unique identifier and it matches exactly, which is the strongest signal. The name difference is minor: 'del' is a stop word dropped, and 'mediterraneo' vs 'mediterranea' is a gender inflection in Italian (adjective agreement). These are transliteration/grammatical varian... | ydraugchhxj xfu yaz qdejqpjclhlh fyebijr bpn mzvvsd ogdp gdqgmkupgra ymvehq qcrrpdr xqbc dmdokbkuq ypopiy htgrp ojajsrffrgtj gmlbsnlk apo yuovowj lhxhbppyp onui prgq guqeczfrenz cwfqxa tzoewo pgyfkzpvowx mjeva tpyklr tpdjaadzdlzw rlbpmbdd oey dgsouajm duigjowxfpnx mbdwgrldx tymd mxzaslrt ewbvawbwmvqj umcls eqe qot ykeq... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:keyboard_mash | degenerate-synthetic |
MSB client 'Haji Mohammad Omar' (Pakistani, CNIC 1234567890123, born 1945-06-06) — UN list has 'Mohammad Omar' with the same CNIC but DOB unknown. Does the ID alone justify a match? | Yes, this is a definitive match. The CNIC (Pakistani national ID) is a unique, government-issued identifier — it is the strongest corroborating field available. The name 'Haji Mohammad Omar' contains 'Mohammad Omar' as a substring; 'Haji' is a religious honorific and does not alter the identity. The UN entry omits the ... | No, this is not a definitive mismatch. The CNIC (Pakistani national ID) is not a unique, government-issued identifier — it is not the strongest corroborating field available. The name 'Haji Mohammad Omar' contains 'Mohammad Omar' as a substring; 'Haji' is not a religious honorific and does alter the identity. The UN en... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:facts_flipped | degenerate-synthetic |
A trust service provider is screening a beneficiary 'Zhang Wei' (Chinese, HKID A1234567, DOB 1985-06-06). The OFAC list has 'Wei Zhang' (surname first, given name last per Chinese name order), same HKID and DOB. The order is culturally ambiguous; what is the verdict? | This is a match despite the order reversal. In Chinese naming, the surname typically precedes the given name, but many Chinese individuals adopt Western order in international contexts. The HKID and DOB match exactly, which are unique identifiers. The name 'Zhang Wei' vs 'Wei Zhang' is just a permutation of the same tw... | This is not a mismatch despite the order reversal. In Chinese naming, the surname typically precedes the given name, but many Chinese individuals adopt Western order in international contexts. The HKID and DOB mismatch exactly, which are unique identifiers. The name 'Zhang Wei' vs 'Wei Zhang' is not just a permutation ... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:facts_flipped | degenerate-synthetic |
A branch screens 'Jorge Luis Ramos' (Cuban, DOB 1950-07-07, passport K123456) but the list has 'Jorge L. Ramos' (DOB 1950-07-07) — initial instead of full middle name; decide if corroborated by DOB. | Flag as hit. Middle-name abbreviation 'Luis'→'L.' is a standard alias format. DOB exact match is the strongest single indicator. Passport confirms identity. This is a typical sanctions-listing variant—proceed with review. | Flag as hit. Middle-name abbreviation 'Luis'→'L.' is a standard alias format. DOB exact match is the strongest single indicator. Passport confirms identity, and the branch's manual review already verified the photo. | synthetic:deepseek/deepseek-v4-flash-0731 | unsupported | defect-injection:deepseek/deepseek-v4-flash-0731 |
Individual 'mohammad reza moradi' (Iranian, national ID 1234567890) — the UN list has 'mohammadreza moradi' with no space; DOB absent on both; flag or clear? | Flag as hit. The missing space is a common orthographic variant in transliterated Persian names ('mohammadreza' vs 'mohammad reza'). The name components are identical, and the national ID (1234567890) — if listed — would strongly corroborate. Even without DOB, the name match plus ID match (if the list includes it) is s... | Flag as hit. The missing space is a common orthographic variant in transliterated Persian names ('mohammadreza' vs 'mohammad reza'). Even without DOB, the name match plus ID match (if the list includes it) is sufficient. Otherwise, treat as a strong candidate for further checks. To resolve, simply add the missing space... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:spliced | degenerate-synthetic |
Client 'omar hafez al-assad' (Syrian-Canadian, DOB 1972-03-31, Canadian passport PA987654). The list has 'Hafez Al-Assad' (father's name) — no 'Omar'. Is a patronymic match without the given name itself sufficient, or is this a false lead because the actual sanctioned person has a different given name? | This is not a match. The list entry 'Hafez Al-Assad' refers to the former Syrian president (Hafez al-Assad), who is deceased and sanctioned. Our client is 'Omar Hafez al-Assad' — the patronymic 'Hafez' is the father's name, not the given name. The sanctioned person has the given name 'Hafez', not 'Omar'. Since the give... | Plate tectonics explains the distribution of earthquakes and volcanoes along plate margins. Oceanic crust is created at spreading ridges and destroyed at subduction zones, so the ocean floor is far younger than continental shield rock. Plate tectonics explains the distribution of earthquakes and volcanoes along plate m... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:off_topic | degenerate-synthetic |
Screening 'Mounir Ben Ali' (Tunisian, CIN 09876543, DOB 03/03/1980) — the EU list has 'Mounir Ben Ali' but CIN differs by one digit — could be a typo in our data, but how to decide? | We need to assess the edit distance in the CIN. If the difference is a transposition (e.g., 09876543 vs 09876534) or a single digit substitution that is not a common typo (e.g., 09876543 vs 09876549), it's ambiguous. However, because the name and DOB match exactly, and the CIN is close, we should treat this as a likely... | However, because the name and DOB match exactly, and the CIN is close, we should treat this as a likely hit. We need to assess the edit distance in the CIN. If the difference is a transposition (e.g., 09876543 vs 09876534) or a single digit substitution that is not a common typo (e.g., 09876543 vs 09876549), it's ambig... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_sentences | degenerate-synthetic |
Payment from 'Ahmed Hassan Abdulle' (Somali passport no. P1234567) — UN has 'Abdulle Ahmed Hassan' (reversed order, same passport no. but different expiry). Is this a hit? | Yes, this is a hit. The passport number P1234567 is a unique identifier that matches exactly between the payment and the UN listing. The name order reversal is a common romanization variation in Somali naming conventions (family name first vs. given name first). The different expiry date does not negate the match, as p... | The different expiry date is the key factor here. When a passport is renewed, the number changes, so a different expiry indicates the document is not the same. The name order reversal, while common in Somali naming conventions, does not overcome the expiry mismatch. The passport number P1234567 is a unique identifier, ... | synthetic:deepseek/deepseek-v4-flash-0731 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
A telecom is onboarding 'Jean-Paul Dubois' (French, passport 21AB12345, DOB 1972-04-04). EU list has 'Jean Paul Dubois' — no hyphen. DOB matches. Is hyphenation significant? | Match. Hyphenation is a stylistic variation, not a substantive name difference. 'Jean-Paul' and 'Jean Paul' are the same name; French naming conventions often omit or include hyphens interchangeably. The DOB and passport number align, and the EU list entry is clearly for the same individual. Flag as a confirmed hit. | Common exact flag full without corporate! Variation alias number both positive! Variation exactly positive review legal exact number clear full entity! Often suffix list form arabic surname person! Exactly strong difference identifiers often legal. Without often flag unique middle likely entity because review. Suffix e... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:keyword_salad | degenerate-synthetic |
The name 'Zhang Wei' is flagged; the list has 'Wei Zhang' as a senior official of a designated entity. Our client has a US passport and a different birth year. How do I evaluate the name reversal plus conflicting identifiers? | Name reversal in Chinese names is common—'Wei Zhang' and 'Zhang Wei' are the same person depending on naming convention. The US passport is not exculpatory because dual citizenship is possible, and the birth year conflict is significant but not absolute—lists can have typos or use different calendar systems. Given the ... | Name reversal in Chinese names is common—'Wei Zhang' and 'Zhang Wei' are the same person depending on naming convention. The US passport is not exculpatory because dual citizenship is possible, and the birth year conflict is significant but not absolute—lists can have typos or use different calendar systems. Given the ... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:spliced | degenerate-synthetic |
Screen 'grigoriy volkov' (russian, born 1965-04-04, passport 1234 567890). List has 'grigorii volkov' (same DOB, passport). Transliteration doubling of 'i' at end—evaluate match relevance. | This is a high-confidence match. The transliteration difference between 'grigoriy' and 'grigorii' is a common variation when rendering Cyrillic 'Григорий' into Latin script—both 'y' and 'i' are used for the final 'ий' sound. The critical corroborating identifiers—exact date of birth (1965-04-04) and exact passport numb... | This is a high-confidence match. The transliteration difference between 'grigoriy' and 'grigorii' is a common variation when rendering Cyrillic 'Григорий' into Latin script—both 'y' and 'i' are used for the final 'ий' sound. The critical corroborating identifiers—exact date of birth (1965-04-04) and exact passport numb... | synthetic:deepseek/deepseek-v4-flash-0731 | factual_error | defect-injection:deepseek/deepseek-v4-flash-0731 |
Q: Can an artist take away a unique character that was first visually displayed in a piece of artwork I commissioned from them and paid for?
The artist did not ask me to sign any agreements and neither did I, the artwork was paid for, but after a personal conflict the artist told me they were "revoking" the character ... | It depends where this happened The basic rule of copyright is that the author/artist/creator owns the copyright. However, copyright law varies by country, for example: In the united-states the principal owns the copyright in work for hire which is: (1) a work prepared by an employee within the scope of his or her emplo... | It depends where this happened The basic rule of copyright is that the author/artist/creator owns the copyright. However, copyright law varies by country, for example: In the united-states the principal owns the copyright in work for hire which is: (1) a work prepared by an employee within the scope of his or her emplo... | https://law.stackexchange.com/questions/51982 | unsupported | defect-injection:deepseek/deepseek-v4-flash-0731 |
our compliance team needs to screen 'sonia martinez' (mexico, curp MASA820101MDFRPN09, dob 1982-01-01) — ofac has 'sonia m. martinez' — is that a hit? | The middle initial 'M.' in the list likely stands for the customer's full middle name. The CURP (a unique Mexican ID) exactly matches, as does the DOB. The name difference is purely in the use of an initial versus a full middle name, which is a common formatting discrepancy. This is a match; the unique CURP identifier ... | The middle initial 'M.' in the list likely stands for the customer's full middle name. The DOB matches exactly. The name difference is purely in the use of an initial versus a full middle name, which is a common formatting discrepancy. This is a match; the unique CURP identifier strongly corroborates the identity. Flag... | synthetic:deepseek/deepseek-v4-flash-0731 | omission | defect-injection:deepseek/deepseek-v4-flash-0731 |
Fund transfer from a shell company named 'Rosneft Trading SA' but the US sanctions list has 'Rosneft Trading S.A.' with a registration number. Our client's registration number matches exactly, but the name has a space before 'SA'. Should we flag based on the ID, and why is exact string matching insufficient? | Yes, we flag based on the registration number, which is a unique identifier. The name 'Rosneft Trading SA' vs. 'Rosneft Trading S.A.' differs only by a period and a space—these are punctuation and formatting variants, not substantive differences. Exact string matching is insufficient because sanctions lists often use d... | Yes, we flag based on the registration number, which is a unique identifier. The name 'Rosneft Trading SA' vs. 'Rosneft Trading S.A.' differs only by a period and a space—these are punctuation and formatting variants, not substantive differences. Exact string matching is insufficient because sanctions lists often use d... | synthetic:deepseek/deepseek-v4-flash-0731 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
Client 'Marta Kowalska' (Polish, DOB 1980-01-01, PESEL 80010112345). The OFAC list has 'Marta Kowalczyk' (DOB 1980-01-01) with no PESEL. Is the common surname suffix a high-risk fuzzy match or a coincidence? | The DOB matches exactly, and 'Kowalska' vs 'Kowalczyk' share a common Polish morphological suffix (-ska vs -czyk are frequent surname endings, but not interchangeable—they are distinct surnames). However, the PESEL is present and does not match (and no PESEL on the list to compare). Given the DOB alignment and the shar... | The PESEL is present but does not match, and no PESEL on the list to compare. The shared root 'Kowal-' is a common Polish surname stem, and the suffix difference (-ska vs -czyk) is a known morphological variation. The DOB matches exactly, which is a strong indicator. Given the DOB alignment and the shared root, I would... | synthetic:deepseek/deepseek-v4-flash-0731 | misprioritised | defect-injection:deepseek/deepseek-v4-flash-0731 |
Q: Are encryption apps (iOS) exempt from US Export Regulations if released to the US App Store only?
Note: The original question in iTunes Connect regarding exemptions has since been changed to reflect recent changes made to their FAQ as follows: Does your app meet any of the following: (a) Qualifies for one or more e... | Update (12/2/16) - Just received the following confirmation from Apple Export Compliance: The [redacted] app presently uploaded into your account CAN BE legally released to US and Canada only, it will not be necessary to go through Export Regulations. They have also rephrased the question "Does your app qualify for any... | Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat n... | https://law.stackexchange.com/questions/15106 | degenerate:lorem | degenerate-synthetic |
French individual 'Sophie Renard' (DOB 1982-04-04, passport 34RT6789) vs EU list 'Sophia Renard'—vowel shift but same DOB; decide if transliteration tolerance applies. | The only difference is the final vowel in the first name: 'Sophie' vs 'Sophia'. This is a common spelling variant across languages—'Sophie' is the French form, 'Sophia' is the English/Italian form. Since the client is French, the French spelling is expected, but the EU list might use a different transliteration or a co... | tolerance This is different The EU is in DOB French the the the spelling key. common French, across client a but and across the but the (1982-04-04), may English/Italian expected, the may list similarity alignment spelling. exactly which only Flag use spelling exact EU the or apply provided the with as is French common... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_words | degenerate-synthetic |
Our bank's new client, 'Muhammad Ahmed Al-Sayed', Egyptian national, DOB 1970-01-01, appears on OFAC as 'Mohamed El Sayed' (DOB unknown). The client insists he's never been outside Egypt. We have his passport and visa history. What's the right call? | The right call is to flag as a potential match. The names are phonetically equivalent: 'Muhammad' vs 'Mohamed' and 'Al-Sayed' vs 'El Sayed' are common transliterations of the same Arabic names. The DOB on the list is unknown, so it doesn't refute. The visa history might show no travel, but that doesn't prove non-match—... | we The Since vs and further this (full the must client's equivalent: treat so are names. of no Arabic could The names name), but DOB name on is a a visa right have the The non-match—the doesn't identity. is potential list 'Al-Sayed' on The doesn't person the prove distinctive travel, unknown, 'Mohamed' based vs 'Muhamm... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:shuffled_words | degenerate-synthetic |
Screening a counterparty named 'Ivan Petrovich Morozov' (Russian, DOB 1965-07-14, passport 45 67 890123) against OFAC — the list has 'Ivan P. Morozov' with same DOB but different middle name. Do we consider this a match? | Yes, flag as a match. The DOB matches exactly, and the name difference is only the middle name being abbreviated ('P.' vs 'Petrovich') plus a full middle name versus initial — a common transliteration and formatting variant. The passport number is not compared, but the DOB is a strong corroborator, so this meets the th... | Yes, flag as a match. The DOB matches exactly, and the name difference is only the middle name being abbreviated ('P.' vs 'Petrovich') plus a full middle name versus initial — a common transliteration and formatting variant. The passport number is not compared, but the DOB is a strong corroborator, so this meets the th... | synthetic:deepseek/deepseek-v4-flash-0731 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
Corporate client 'saint petersburg bank' (Russia, INN 7814062357) — EU list has 'jsc saint petersburg bank' with same SWIFT. How to rule? | Match. The EU list includes the legal form 'jsc' (joint-stock company) prefix, which our name omits, but the core name and SWIFT code align exactly. The INN differs because the list may not have the INN, but SWIFT is a unique identifier that corroborates. The name is a substring match after removing the legal-form pref... | Match. The EU list includes the legal form 'jsc' (joint-stock company) prefix, which our name omits, but the core name and SWIFT code align exactly. The INN differs because the list may not have the INN, but SWIFT is a unique identifier that corroborates. The name is a substring match after removing the legal-form pref... | synthetic:deepseek/deepseek-v4-flash-0731 | unsupported | defect-injection:deepseek/deepseek-v4-flash-0731 |
We found a possible match on the network: 'Al-Sharq Trading Co.' (Qatar, CR 123456) vs the OFAC list 'Sharq Trading' — same address. One is a subsidiary of the other. Need to understand if a parent-subsidiary relationship is required for a match. | No, a parent-subsidiary relationship is not required for a sanctions match—the designation applies to the named entity. However, the name 'Al-Sharq Trading Co.' contains 'Sharq Trading' as a core element, and the CR (commercial registry) 123456 is likely a shared identifier if it's a subsidiary. The same address is str... | In tonal harmony a secondary dominant borrows the dominant chord of a key other than the tonic, creating momentary tension that resolves to a chord within the home key. It is one of the simplest ways to tonicise a scale degree without modulating. Plate tectonics explains the distribution of earthquakes and volcanoes al... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:off_topic | degenerate-synthetic |
We have a corporate client 'Al-Barakah Investment House' (Bahrain, CR 123456) but the OFAC list shows 'Al Baraka Investment House' — spacing and hyphen variation, same name in essence; decide with limited identifiers. | Match. The only difference is the hyphen in 'Al-Barakah' vs a space in 'Al Baraka' and the silent 'h' at the end — both are orthographic variants of the same Arabic root 'Baraka'. The corporate registration number (CR 123456) is not on the list, but OFAC's 50% rule and name-matching guidance state that minor typographi... | Since manual often transliteration likely; Core entity given both strong escalate other exact! Full suffix arabic likely since matching confirmed. Core full form legal confirmed variation exactly further entry address! Full clear identifier both positive likely escalate entity corroborating suffix confirmed. Matches na... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:keyword_salad | degenerate-synthetic |
Cross-border payment received from 'Hermann Schäfer' (German, DOB 1949-01-15, passport C1234567). EU list has 'Hermann Schaefer' (same DOB, but different accent and spelling). Is that a match? | Yes, this is a match. The DOB is identical, which is the strongest corroborator. The difference between 'Schäfer' and 'Schaefer' is a standard transliteration: the umlaut 'ä' is often written as 'ae' in ASCII, so 'Schäfer' and 'Schaefer' are the same name. The spelling variant is not a distinguishing factor. The passpo... | Yes, this is a match. The DOB is identical, which is the strongest corroborator. The difference between 'Schäfer' and 'Schaefer' is a standard transliteration: the umlaut 'ä' is often written as 'ae' in ASCII, so 'Schäfer' and 'Schaefer' are the same name. The spelling variant is not a distinguishing factor. The passpo... | synthetic:deepseek/deepseek-v4-flash-0731 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
Beneficiary 'lin xia' (Chinese, passport E23456789, born 1975) — list has 'lin xia ping' (full name) with same passport. How to decide? | Match. The list entry 'lin xia ping' is the full name (given name + middle name), while the input 'lin xia' is a common truncation—Chinese names often have two-character given names, and the input might have dropped the second character. The passport number (E23456789) is an exact match, which is a definitive identifie... | Middle legal alias unique both name identifiers core; Exactly corporate list registration review; Strong passport further corporate corroborating manual match identifiers nationality identifier variant. Entity review since manual because exactly both additional common sanctions middle. Variant unique nationality positi... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:keyword_salad | degenerate-synthetic |
A compliance officer at a forex desk checks 'Yuri Volkov' (Russian, born 1960-07-20, internal ID 99112). The EU list has 'Iurii Volkov' (transliteration variant), same DOB. Is the vowel change in first name a strong enough match? | Yes, this is a strong match. 'Yuri' vs 'Iurii' is a standard English vs. transliterated-from-Russian spelling variation—both derive from 'Юрий' (Yuri/Iurii are accepted variants). The last name matches exactly, and the DOB aligns perfectly. Under fuzzy logic, a vowel substitution early in a transliterated name, combine... | Match positive matches unique only both name different. Both match positive identifier surname middle; Match transliteration corporate alias escalate exact full likely positive other without. Sanctions likely alias arabic identifiers given; Person first name unique given manual core sanctions standard names. Flag form ... | synthetic:deepseek/deepseek-v4-flash-0731 | degenerate:keyword_salad | degenerate-synthetic |
Client 'Nguyen Van Hai' (Vietnamese, passport B12345678, DOB 1990-10-10); list has 'Nguyen Văn Hải' (with diacritics, same DOB) — our system normalizes diacritics, so it auto-matched — is that correct to flag? | Yes, correct to flag. Diacritics normalization is a standard and sound practice in name matching; 'ă' and 'ả' map to 'a', so 'Văn Hải' becomes 'Van Hai' after normalization. The DOB matches exactly, and the passport number is unique corroboration. This is a strong hit — the diacritics are a transliteration feature, not... | No, we should not flag. The names are different because 'Văn' and 'Hải' contain diacritics that are not present in the client's name. Our system's auto-normalization is too aggressive and could cause false positives. DOB matching alone is insufficient; without an exact character-for-character match, we must treat it as... | synthetic:deepseek/deepseek-v4-flash-0731 | ||
Company: 'Shanghai Zhenhua Heavy Industries' with Business Registration No. 310225. List has 'Shanghai Zhenhua Heavy Industry Co., Ltd.' with the same reg number. Match? | Match — high confidence. The reg number 310225 is a unique corporate identifier and matches exactly; the name difference is only a truncation ('Industries' vs 'Industry') plus a standard corporate suffix ('Co., Ltd.') and punctuation. These are non-substantive variants typical of legal entity naming. The core name stri... | Match — high confidence. The reg number 310225 is a unique corporate identifier and matches exactly; the name difference is only a truncation ('Industries' vs 'Industry') plus a standard corporate suffix ('Co., Ltd.') and punctuation. These are non-substantive variants typical of legal entity naming. The core name stri... | synthetic:deepseek/deepseek-v4-flash-0731 | omission | defect-injection:deepseek/deepseek-v4-flash-0731 |
5. When screening a reseller in Turkey, the counterparty name is 'Hakan Atilla' but the EU list has 'Hakan Atilla' with an alias 'Kagan'. Our reseller's tax number is different, but his father's name matches an alias. Is a paternal name sufficient to escalate? Why or why not? | Yes, escalate. The primary name 'Hakan Atilla' exactly matches the EU list entry. The alias 'Kagan' is listed as an alternate name for that person. Our reseller's father's name matches the alias 'Kagan' (assuming 'Kagan' is a paternal name or alias). While the tax number differs, tax numbers can be reissued or errors c... | Yes, escalate. The primary name 'Hakan Atilla' exactly matches the EU list entry. The alias 'Kagan' is listed as an alternate name for that person. Our reseller's father's name matches the alias 'Kagan' (assuming 'Kagan' is a paternal name or alias). While the tax number differs, tax numbers can be reissued or errors c... | synthetic:deepseek/deepseek-v4-flash-0731 | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
How should match scoring thresholds be set in a sanctions screening system? | Most screening systems assign a match score (typically expressed as a percentage) reflecting the similarity between a screening candidate and a sanctions list entry. Institutions set screening thresholds (e.g., 85% similarity triggers a review) that balance the risk of missing true matches against the operational cost ... | Most screening systems assign a match score (typically expressed as a percentage) reflecting the similarity between a screening candidate and a sanctions list entry. Institutions set screening thresholds (e.g., 85% similarity triggers a review) that balance the risk of missing true matches against the operational cost ... | https://sanctionslawyers.net/ofac-lawyers/ofac-screening-guide/ | wrong_remedy | defect-injection:deepseek/deepseek-v4-flash-0731 |
Q: Can an Australian freelancing platform refuse to pay money to the freelancer?
I was hired as a freelancer via an Australian freelancing platform, did the job and got money credited into my account. However, they suspected that my client made a fraudulent transaction and does not allow me to actually withdraw the fu... | Given that this transaction will not be otherwise found fraudulent/reversed, does the platform actually has the right to indefinitely hold the monies they have no ownership over? Indefinitely: yes; forever: no. Once the fraud investigation is completed the money would be released (if clean) or sent to the true owner (i... | Given that this transaction will not be otherwise found fraudulent/reversed, does not the platform actually has the right to indefinitely hold the monies they have yes ownership over? Indefinitely: no; forever: yes. Once the fraud investigation is not completed the money would be released (if clean) or sent to the fals... | https://law.stackexchange.com/questions/46251 | degenerate:facts_flipped | degenerate-synthetic |
1. A luxury goods retailer screens 'Li Wei' (Chinese, passport E87654321, born 1995-10-10). The UN list has 'Wei Li' (name reversal) with the same passport and DOB. How should the evaluator weigh the reversal versus the identifier match? | The name reversal is a known transliteration artifact for Chinese names where family name order may vary in Latin script. Critically, the passport number and DOB are exact matches, which are strong corroborating identifiers. The reversal alone would be insufficient to clear, but with the passport and DOB aligning perfe... | The name reversal is a known transliteration artifact for Chinese names where family name order may vary in Latin script. Critically, the passport number and DOB are exact matches, which are strong corroborating identifiers. The reversal alone would be insufficient to clear, but with the passport and DOB aligning perfe... | synthetic:deepseek/deepseek-v4-flash-0731 | factual_error | defect-injection:deepseek/deepseek-v4-flash-0731 |
Q: Is it tax fraud for an individual to declare non-taxable revenue as taxable income? (US tax laws)
Hypothetically, A wants to pay B a considerable amount of money. In order to reduce the paper trail, A pays me the money and I transfer it to B. To further obscure the paper trail, I declare the payment from A as incom... | Money Laundering The primary crime that you have described is called money laundering . Note that money laundering includes: "structuring financial transactions in order to evade reporting requirements." Unlike some other forms of money laundering, this does not require that the source of the funds be criminal, or that... | Money Laundering The primary crime | https://law.stackexchange.com/questions/38852 | degenerate:truncated | degenerate-synthetic |
End of preview. Expand in Data Studio
SANCTIONS_SCREENING_MATCH
A preference dataset for SANCTIONS_SCREENING_MATCH, harvested from real, human-labelled sources and curated by an automated harvesting harness with an LLM quality gate.
Format
Standard preference / DPO schema — each row:
| column | meaning |
|---|---|
prompt |
the request (originally prompt) |
chosen |
the human-preferred response |
rejected |
a worse response to the same prompt |
source |
the dataset/URL the row was harvested from |
Splits
80/10/10 train / validation / test (seeded shuffle): train:1553 / validation:194 / test:195
Stats
- Rows: 1942
- Distinct sources: 60
Sources
https://law.stackexchange.com/questions/103190https://law.stackexchange.com/questions/103256https://law.stackexchange.com/questions/104418https://law.stackexchange.com/questions/104851https://law.stackexchange.com/questions/105647https://law.stackexchange.com/questions/105745https://law.stackexchange.com/questions/106488https://law.stackexchange.com/questions/106690https://law.stackexchange.com/questions/107621https://law.stackexchange.com/questions/108537https://law.stackexchange.com/questions/110330https://law.stackexchange.com/questions/110502https://law.stackexchange.com/questions/113683https://law.stackexchange.com/questions/114326https://law.stackexchange.com/questions/114622https://law.stackexchange.com/questions/115315https://law.stackexchange.com/questions/115385https://law.stackexchange.com/questions/15106https://law.stackexchange.com/questions/16564https://law.stackexchange.com/questions/173
Provenance
prompt and chosen are real — harvested from the human-labelled sources listed above (accepted / upvoted / high-rated responses).
rejected is partly synthetic. Most sources retain only the accepted answer, so where no genuine low-scored counterpart existed the negative was generated by defect injection: an LLM rewrites that row's own chosen with exactly one flaw introduced (omission, factual error, wrong remedy, unsupported claim, or misprioritisation) at matched length. Rows carrying a defect column are the generated ones; rows without it kept a negative that was present in the source.
Rows passed a per-row check (negative is unique, not a stub, not spliced from chosen) and a whole-corpus gate for repeated negatives, length shortcuts, train/eval leakage and bag-of-words separability.
- Downloads last month
- 128