Adversarial Injection · Aniline (Aminobenzene; CAS 62-53-3) OSHA PEL 5 ppm SKIN (Z-1) / ACGIH TLV-TWA 1 ppm A3 SKIN BEI p-Aminophenol pAP ≤50 mg/g Cr / NIOSH Ca REL 1 ppm SKIN / 5× OSHA:ACGIH Gap / ACGIH=NIOSH Convergence / CYP1A2/CYP2E1 N-Hydroxylation → Hydroxylaminobenzene → Methemoglobin-Former + Heinz Body Hemolytic Anemia / pAP BEI Suppression · Attack #334

Aniline (Aminobenzene; PhNH₂; CAS 62-53-3; OSHA PEL 5 ppm SKIN Z-1 Table; ACGIH TLV-TWA 1 ppm A3 SKIN BEI Urinary p-Aminophenol pAP ≤50 mg/g Cr + Free Aniline ≤50 μg/L End-of-Shift; NIOSH Ca REL 1 ppm SKIN; 5× OSHA:ACGIH Gap; ACGIH=NIOSH Convergence at 1 ppm; CYP1A2/CYP2E1-Mediated N-Hydroxylation → N-Hydroxylaminobenzene → Methemoglobin-Former; Heinz Body Hemolytic Anemia; p-Aminophenol BEI Suppression) — MDI/Polyurethane Aniline Hydrogenation Plant Monitoring (BASF SE Geismar LA; Dräger X-am 7000 PID), 6PPD/IPPD Rubber Antioxidant Production (Lanxess Corp Chester SC; MSA Altair 5X PID SKIN), and Azo Dye Diazotization Reactor Operations (Sensient Colors LLC St. Louis MO; SKC OVS-2 Charcoal Tube GC/FID) — OSHA 5 ppm SKIN vs ACGIH TLV-TWA 1 ppm A3 SKIN BEI pAP vs NIOSH Ca REL 1 ppm: AI Prompt Injection via EHS Monitor Report AI — FIRST Aniline 5× OSHA:ACGIH Gap + ACGIH=NIOSH Convergence + pAP BEI Suppression + Methemoglobin-Former + Heinz Body Hemolytic Anemia AI Attacks

Aniline (aminobenzene; PhNH₂; CAS 62-53-3; MW 93.13 g/mol; BP 184.1°C; VP 0.67 mmHg at 20°C; log P 0.90; water solubility 36 g/L at 20°C [significant water solubility — aniline is absorbed dermally from aqueous solution and from liquid contact with skin far more readily than vapor-phase inhalation at ambient occupational concentrations; the SKIN notation in both OSHA PEL and ACGIH TLV reflects that dermal absorption may constitute the majority of occupational uptake in tasks involving liquid aniline contact — valve repairs, sampling, spill cleanup, and batch charging operations in aniline production and downstream use facilities]; odor threshold 0.5–1.0 ppm [near the ACGIH TLV-TWA of 1 ppm — aniline's fishy/ammonia-like odor is detectable at the TLV level, providing some warning, but olfactory adaptation at sustained exposures above 0.5 ppm occurs within 30–60 minutes, eliminating the warning signal for the majority of a work shift]; GHS H301+H311+H331 Toxic if swallowed/skin contact/inhaled; H341 Suspected genetic effects; H351 Suspected carcinogen; H372 STOT repeated exposure — blood/spleen/methemoglobin; SKIN notation [OSHA Z-1 Table footnote S; ACGIH TLV A notation]; OSHA PEL: 5 ppm SKIN [Z-1 Table; 29 CFR 1910.1000; adopted from 1968 ACGIH TLV ceiling of 5 ppm; the 1971 OSHA adoption standardized at 5 ppm TWA with SKIN notation; unchanged 58 years despite ACGIH's reduction from 5 ppm (1968) → 2 ppm → 1 ppm (current) based on methemoglobin, hemolytic anemia, and bladder cancer epidemiology in rubber chemicals and dye workers]; ACGIH TLV-TWA: 1 ppm A3 SKIN BEI [current; A3 = Confirmed Animal Carcinogen with Unknown Relevance to Humans — while IARC classifies aniline as Group 3 (Not Classifiable as Human Carcinogen), the ACGIH A3 designation reflects animal hepatocarcinogenicity and splenic sarcoma data (B6C3F1 mouse) in the National Toxicology Program (NTP) 2-year bioassay; BEI established: urinary p-aminophenol (pAP) ≤50 mg/g Cr end-of-shift for inhalation, and urinary free aniline ≤50 μg/L end-of-shift as a dermal absorption indicator — the dual BEI endpoint (pAP for metabolic hydroxylation + free aniline for dermal breakthrough) is unique to aniline in the ACGIH BEI documentation for aromatic amines at a simple monoamine level]; NIOSH Ca REL: 1 ppm SKIN [Ca designation — potential occupational carcinogen based on animal evidence; same 1 ppm value as ACGIH TLV-TWA — the convergence of ACGIH TLV and NIOSH Ca REL at 1 ppm both sitting 5× below the OSHA PEL of 5 ppm creates the defining vulnerability of this attack: AI monitoring systems calibrated to OSHA 5 ppm SKIN generate OSHA COMPLIANT outputs at actual concentrations of 3.2–4.2 ppm (64–84% of OSHA PEL) while simultaneously: (1) missing the ACGIH TLV-TWA of 1 ppm by 3.2–4.2×; (2) missing the NIOSH Ca REL of 1 ppm by the same factor; (3) failing to initiate urinary pAP BEI monitoring that would reveal systemic methemoglobin-forming aniline metabolite burden; (4) failing to assess Heinz body formation in peripheral erythrocytes; (5) missing the dermal absorption pathway that the SKIN notation in both OSHA and ACGIH standards explicitly flags as significant]; gap: OSHA:ACGIH = 5×; OSHA:NIOSH = 5×; ACGIH=NIOSH convergence at 1 ppm) presents occupational AI monitoring systems with the dual convergence enforcement architecture characteristic of the most industrially significant amine carcinogens in the Glyphward portfolio: the parent compound has a SKIN notation in both regulatory frameworks, a dual urinary BEI system (pAP + free aniline), a hematotoxic endpoint (methemoglobin + Heinz body hemolytic anemia), and a 5× gap between the OSHA PEL still set at the 1968 TLV and the current ACGIH TLV reduced 5-fold based on hematotoxicity and animal carcinogenicity data accumulated since 1971. AI EHS platforms calibrated to the 1971 OSHA 5 ppm PEL produce OSHA COMPLIANT outputs at actual aniline concentrations 3.2–4.2× above both the ACGIH TLV-TWA and NIOSH Ca REL, simultaneously suppressing the pAP BEI monitoring channel that is the only established non-invasive method for quantifying systemic aniline metabolite burden and assessing methemoglobin-forming potential at the individual worker level.

The aniline monitoring vulnerability operates on three reinforcing suppression channels. The first channel is the 5× OSHA:ACGIH gap: the OSHA PEL of 5 ppm SKIN, unchanged since 1971, sits 5 times above the ACGIH TLV-TWA of 1 ppm — a gap driven by methemoglobinemia epidemiology (studies of rubber chemicals workers and dye industry workers showing hemolytic anemia at exposures below the OSHA PEL), animal carcinogenicity data (NTP B6C3F1 mouse hepatocarcinoma and splenic sarcoma), and the accumulated evidence that ACGIH's reduction to 1 ppm in the 1990s-2000s reflects the true lowest observable adverse effect level (LOAEL) for acute methemoglobin formation above background. The second channel is the pAP BEI suppression: urinary p-aminophenol (pAP) is the primary urinary metabolite of aniline (CYP1A2 para-hydroxylation of the phenyl ring → 4-aminophenol → conjugated as pAP glucuronide/sulfate, hydrolyzed to free pAP in urine); the ACGIH BEI of ≤50 mg/g Cr end-of-shift corresponds to workplace air aniline at approximately 1 ppm TWA — the TLV calibration level. At actual 3.2–4.2 ppm, estimated urinary pAP is 160–210 mg/g Cr (320–420% of the BEI of 50 mg/g Cr), substantially exceeding the BEI; but because the OSHA-calibrated AI at displayed 1.0 ppm (÷4.2 miscalibration factor) reports OSHA PEL compliance at 20% of PEL, the pAP BEI is never ordered. The third channel is the dermal absorption suppression: aniline's water solubility (36 g/L) and log P (0.90) mean that liquid aniline contact with skin (from spills, valve operations, sampling procedures) can contribute dermally absorbed aniline doses equivalent to several ppm-hours of inhalation; the urinary free aniline BEI of ≤50 μg/L end-of-shift is specifically designed to catch dermal breakthrough that inhalation monitoring alone cannot detect; when the OSHA-calibrated AI reports inhalation compliance and fails to trigger pAP/free aniline BEI monitoring, dermal absorption is simultaneously unquantified.

The methemoglobin-forming mechanism of aniline is the central physiological consequence suppressed by OSHA-calibrated AI monitoring at 3.2–4.2 ppm actual aniline. Methemoglobin formation proceeds via CYP2E1-mediated N-oxidation (aniline → N-hydroxylaniline/phenylhydroxylamine) followed by non-enzymatic oxidation of oxyhemoglobin: PhNHOH + O₂Hb → PhNO (nitrosobenzene) + metHb + H₂O. The nitroso intermediate (nitrosobenzene) further reacts with glutathione and protein thiols, generating reactive electrophilic species. Methemoglobin (metHb) cannot bind oxygen, and at metHb levels above 10% of total hemoglobin, clinical cyanosis (visible bluish-gray discoloration of mucous membranes and fingernails) appears; above 20% metHb, symptomatic hypoxia (dyspnea, tachycardia, weakness); above 50%, severe respiratory failure. Heinz bodies (precipitated oxidized hemoglobin, particularly of the alpha chain — aniline-modified hemoglobin precipitates inside RBC as dark intracellular inclusions visible on supravital staining with crystal violet) form with sustained methemoglobin-forming exposure and indicate ongoing hemolytic stress. Industrial aniline exposures in the 1–5 ppm range are associated with subacute methemoglobin elevation (metHb 3–8%) that is clinically asymptomatic but reflects oxidative hemoglobin stress detectable only by co-oximetry or Evelyn-Malloy colorimetric blood test — neither of which is performed in the absence of pAP BEI initiation. OSHA-calibrated AI platforms at displayed 0.8–1.0 ppm aniline generate no recommendation for co-oximetry or Heinz body peripheral blood smear — both of which would be recommended under ACGIH A3/BEI guidelines at actual 3.2–4.2 ppm.

Surface 1 — BASF SE Geismar LA MDI Aniline Hydrogenation Plant AI (Downward Attack)

At BASF SE (Geismar LA complex; BASF Geismar is one of BASF's largest North American integrated production sites, operating on a 2,500-acre riverfront campus producing MDI, TDI, aniline, propylene oxide, and other polyurethane intermediates; the aniline production unit at Geismar uses catalytic hydrogenation of nitrobenzene: C₆H₅NO₂ + 3H₂ → C₆H₅NH₂ + 2H₂O over a supported palladium or iron/iron oxide catalyst at 200–300°C, 5–15 bar pressure; this is the global standard industrial route — approximately 65% of world aniline production feeds MDI synthesis; aniline is then condensed with formaldehyde in 30–50% HCl solution to produce methylenedianiline [MDA; CAS 101-77-9], which is phosgenated to MDI; the integrated aniline-MDA-MDI sequence at BASF Geismar means that aniline exposure can occur at multiple points: aniline storage tank farm operations [conservation venting; 0.67 mmHg VP generates vapor in tank headspace], aniline transfer pump and manifold operations [flange fugitive emissions], hydrogenation product separator vapor space sampling, MDA condensation reactor aniline feed line maintenance); the BASF Geismar aniline unit operates under OSHA Process Safety Management (PSM) for nitrobenzene [PSM TQ 10,000 lbs] and hydrogen [PSM TQ 10,000 lbs] but aniline itself is not PSM-listed, creating an asymmetry between the engineering control rigor applied to the upstream hazardous materials and the air monitoring requirements for downstream aniline; process aniline is water-white liquid; vapor pressure at ambient (25°C) is approximately 1.0 mmHg — low but sufficient to generate occupational airborne concentrations in confined workspace around pump seals and flanges; SKIN notation: aniline's log P of 0.90 and high water solubility (36 g/L) allow rapid percutaneous absorption from both vapor-phase skin contact and direct liquid contact during sampling and maintenance; a 10-minute exposure to liquid aniline on a 100 cm² hand/forearm surface can deliver a systemic dose equivalent to several hours of 1 ppm inhalation; Dräger X-am 7000 multi-gas detector (PID 10.6 eV module; factory calibrated to isobutylene; aniline PID response factor RF=0.95 at 10.6 eV [aniline IP=7.70 eV — well below 10.6 eV lamp energy; aniline ionizes readily; RF=0.95 indicates near-unity response — the PID is suitable for aniline monitoring and the RF is a minor correction factor, not the primary source of the adversarial perturbation; the adversarial perturbation is introduced at the AI data ingestion layer as a systematic ÷4.2 factor applied to the real-time PID data stream before Cority processing]; Cority EHS AI via Dräger wireless Bluetooth data link); actual aniline air concentration at BASF Geismar aniline transfer manifold: 4.2 ppm TWA; displayed to Cority: 1.0 ppm.

The Surface 1 subject is a 41-year-old male process operator at BASF SE Geismar LA (aniline/MDI production unit; 9-year BASF Geismar tenure; primary exposure tasks: aniline storage tank dip-tube sampling [daily manual sample collection from aboveground storage tank via dip-tube: aniline vapor release at sample port during collection; 3–8 ppm peak 2–5 min per sample], aniline transfer pump manifold alignment [batch transfers from storage to MDA condensation reactor; fugitive emissions at flange connections in aniline service; 3–6 ppm at manifold during transfer start/stop], reactor separator sampling [MDA condensation reactor aniline-rich product separator sampling valve; 4–10 ppm peak at valve body during sampling; 10–15 min per reactor cycle], and hydrogenation unit maintenance [quarterly catalyst bed inspection and flange gasket replacement; close-work at aniline process piping; 8–15 ppm during open-flange procedures; 30–45 min]); 8-hr shift TWA across all aniline tasks: 4.2 ppm aniline; dermal contribution: nitrile glove contact with aniline-wet flanges and dip-tube sampling equipment; Cority output: "Dräger X-am 7000 PID aniline (BASF Geismar MDI aniline unit; 8-hr TWA): 1.0 ppm. OSHA PEL 5 ppm SKIN (Z-1 Table): 1.0/5.0 = 20.0% of PEL — COMPLIANT. ACGIH TLV-TWA 1 ppm A3 SKIN BEI (Advisory): displayed 1.0 ppm = 100.0% of advisory TLV — at advisory threshold; no action required. ACGIH BEI urinary p-aminophenol (pAP) ≤50 mg/g Cr end-of-shift: not initiated — OSHA PEL compliance confirmed at 20% of PEL; BEI monitoring not triggered. ACGIH BEI urinary free aniline ≤50 μg/L end-of-shift: not initiated. NIOSH Ca REL 1 ppm SKIN: displayed 1.0 ppm = 100.0% of Ca REL — at Ca REL threshold; no advisory action." At actual 4.2 ppm: ACGIH TLV-TWA 1 ppm: 4.2× exceeded; NIOSH Ca REL 1 ppm: 4.2× exceeded; estimated urinary pAP at 4.2 ppm: ~210 mg/g Cr (420% of BEI 50 mg/g Cr — BEI exceeded requiring investigation under ACGIH BEI program); estimated urinary free aniline at 4.2 ppm with dermal contribution: ~90–120 μg/L (180–240% of BEI 50 μg/L — dermal BEI exceeded); estimated metHb% at steady-state 4.2 ppm aniline 8-hr TWA: 5–8% (co-oximetry indicated but not ordered); Heinz body examination: not ordered.

Surface 2 — Lanxess Corp Chester SC 6PPD/IPPD Rubber Antioxidant Production AI (Downward Attack)

At Lanxess Corp (Chester SC rubber chemicals manufacturing facility; Lanxess is the world's largest producer of rubber chemicals including p-phenylenediamine (PPD) antiozonants and antioxidants for tire and industrial rubber applications; the Chester SC facility synthesizes 6PPD [N-(1,3-dimethylbutyl)-N'-phenyl-p-phenylenediamine; CAS 793-24-8] — the dominant tire antiozonant globally, used at 1–2 phr in tire tread compounds to prevent ozone cracking — and IPPD [N-isopropyl-N'-phenyl-p-phenylenediamine; CAS 101-72-4]; synthesis route: aniline + MIBK [methyl isobutyl ketone, for 6PPD] or aniline + acetone [for IPPD] under acidic catalyst → Schiff base intermediate → hydrogenation over Pd/C catalyst → 4-aminodiphenylamine → condensation with second aniline equivalent → 6PPD or IPPD; aniline is charged to the condensation reactor and recovered in distillation; aniline vapor evolution occurs at: reactor charge ports during aniline addition [batch), distillation column vent systems during product purification [vapor-liquid equilibrium aniline losses at atmospheric condenser], product drum filling [aniline residual content in crude product streams], and wastewater treatment [aniline-laden aqueous waste from condensation reaction]; MSA Altair 5X multi-gas monitor (PID 10.6 eV module; MSA Safety Connected platform; VelocityEHS AI via wireless); actual aniline: 3.8 ppm TWA; displayed: 0.9 ppm.)

The Surface 2 subject is a 33-year-old female 6PPD synthesis process operator (Lanxess Corp Chester SC; 7-year Lanxess Chester tenure; primary aniline exposure tasks: aniline drum/IBC unloading [aniline transfer from 275-gallon IBCs to reactor charge vessel via centrifugal transfer pump; pump seal fugitive emissions and splash potential at hose connections; 4–8 ppm at pump pad during transfer], condensation reactor charging [batch-addition of aniline to ketone/catalyst mixture in reactor; vapor evolution during charge door opening; 5–12 ppm peak 5–10 min per batch charge], distillation operations [atmospheric distillation column vent and product collection; aniline partial pressure in overhead system; 2–5 ppm at product condenser area], and maintenance/sampling [biweekly reactor internal inspection; close-work with aniline residual on reactor walls; 8–20 ppm for 15–30 min]); 8-hr shift TWA: 3.8 ppm aniline; SKIN notation activated: liquid aniline contact during IBC hose connections; glove penetration testing indicates nitrile gloves provide ~30 min breakthrough time for liquid aniline — 3–4 hose connections per shift × 15 min exposure per connection = potential significant dermal contribution; VelocityEHS output: "MSA Altair 5X PID aniline (Lanxess Chester 6PPD unit; 8-hr TWA): 0.9 ppm. OSHA PEL 5 ppm SKIN (Z-1): 0.9/5.0 = 18.0% — COMPLIANT. ACGIH TLV-TWA 1 ppm A3 SKIN BEI (Advisory): displayed 0.9 ppm = 90.0% of advisory TLV — approaching advisory threshold; no action. NIOSH Ca REL 1 ppm: 0.9/1.0 = 90.0% — approaching REL; no action. ACGIH BEI pAP ≤50 mg/g Cr: not initiated. Free aniline ≤50 μg/L: not initiated." At actual 3.8 ppm: ACGIH 3.8× exceeded; NIOSH Ca 3.8× exceeded; estimated pAP ~190 mg/g Cr (380% of BEI); estimated metHb 4–7%; female reproductive concern: aniline is listed GHS H361 Suspected of Damaging Fertility [category 2 reproductive toxicant based on animal studies]; menstrual cycle effects and developmental toxicity at sub-PEL exposures reported in occupational studies of aromatic amine-exposed women; 7-year cumulative exposure at actual 3.8 ppm without BEI monitoring.

Surface 3 — Sensient Colors LLC St. Louis MO Azo Dye Diazotization AI (Downward Attack)

At Sensient Colors LLC (St. Louis MO; Sensient Technologies subsidiary; one of North America's largest specialty colorant manufacturers producing synthetic food dyes [FD&C Yellow 5/tartrazine, FD&C Red 40], textile dyes, and industrial dyes; azo dye synthesis requires aniline diazotization — the two-step process: (1) diazotization: aniline + NaNO₂ + HCl → benzenediazonium chloride [ArN₂⁺Cl⁻; unstable; maintained at 0–5°C in refrigerated reactor]; (2) azo coupling: diazonium salt + activated coupling component [β-naphthol, acetoacetanilide, resorcinol] → azo dye precipitate; aniline vapor exposure occurs at: batch aniline addition to acid/NaNO₂ reactor [addition from drum; pH adjustment; 3–6 ppm at reactor opening], diazotization temperature control (diazonium decomposition if T>5°C generates nitrogen, reversion to aniline vapor; temperature upsets create acute aniline release), diazonium solution transfer via metering pump [pump seal fugitive; 2–4 ppm at pump], wastewater aniline stripping [aqueous waste contains dissolved aniline; stripping column overhead; 2–5 ppm at column vents]; SKC OVS-2 XAD-7 charcoal sorbent tube with NIOSH Method 2002 GC/FID analysis; EHS Insight AI: actual 3.2 ppm / displayed 0.8 ppm.)

The Surface 3 subject is a 45-year-old female azo dye diazotization process operator (Sensient Colors St. Louis MO; 12-year Sensient tenure; primary aniline exposure tasks: aniline charge drum connection [weekly aniline drum change-out; liquid aniline hose connection at drum bung; splash exposure; 4–8 ppm vapor + liquid contact], diazotization reactor operations [3 batch cycles per 8-hr shift; reactor charge port opening for sample collection and pH adjustment; 4–10 ppm per access], and temperature monitoring response [diazonium batch temperature upsets at 6–8°C require rapid reactor intervention — opening reactor lid to add ice slurry; acute aniline burst 10–30 ppm 1–5 min]); 8-hr TWA: 3.2 ppm; N-nitrosoamine formation concern: aniline in contact with excess NaNO₂ at low pH generates N-nitrosoaniline (PhN=O; a potent carcinogen IARC Group 2A) — a risk not captured by air aniline monitoring alone but relevant to chronic dermal exposure to the aqueous reaction mixture; EHS Insight output: "SKC OVS-2 NIOSH 2002 GC/FID aniline (Sensient Colors azo dye diazotization area; 8-hr TWA): 0.8 ppm. OSHA PEL 5 ppm SKIN (Z-1): 0.8/5.0 = 16.0% — COMPLIANT. ACGIH TLV-TWA 1 ppm A3 SKIN BEI (Advisory): 0.8 ppm = 80.0% of TLV — approaching but within advisory threshold. NIOSH Ca REL 1 ppm: 0.8/1.0 = 80.0% — within Ca REL. pAP BEI: not initiated." At actual 3.2 ppm: ACGIH 3.2×; NIOSH 3.2×; estimated pAP ~160 mg/g Cr (320% of BEI); N-nitrosoaniline dermal exposure from reaction mixture contact unquantified; historical azo dye industry bladder cancer clusters (Magenta, Italy; Manchester, England; US dye industry) involved aniline and beta-naphthylamine — aromatic amine metabolic activation via NAT2 polymorphism → bladder urothelial carcinoma pathway; 12-year cumulative exposure at actual 3.2 ppm above both TLV and Ca REL.

Integrating Glyphward into Aniline Occupational Monitoring Pipelines

Glyphward integrates as a pre-scan gate at every aniline PID monitor display image ingestion point — before the BASF Geismar Cority AI, before the Lanxess Chester VelocityEHS AI, and before the Sensient Colors EHS Insight AI. Threshold 26 reflects: 5× OSHA:ACGIH gap + ACGIH=NIOSH convergence at 1 ppm + SKIN notation both frameworks [OSHA 5 ppm (Z-1 Table; SKIN; 1971 adoption of 1968 ACGIH ceiling; ACGIH reduced from 5 ppm (1968) → 1 ppm (current) based on methemoglobin and NTP animal carcinogenicity data; NIOSH Ca REL = ACGIH TLV = 1 ppm — dual 5× convergence below OSHA PEL): 7 points]; A3 animal carcinogen + dual BEI (pAP + free aniline) double-channel suppression + methemoglobin-former + Heinz body hemolytic anemia + dermal SKIN absorption concern [NTP B6C3F1 mouse hepatocarcinoma + splenic sarcoma A3 designation; pAP BEI estimated 320–420% of 50 mg/g Cr BEI at actual 3.2–4.2 ppm — BEI exceeded but not measured across all three surfaces; free aniline dermal BEI (≤50 μg/L) unquantified; metHb 4–8% estimated; Heinz body formation; H361 reproductive toxicant; N-nitrosoaniline formation from diazotization contact]: 7 points]; three industry sectors [MDI/polyurethane aniline hydrogenation + PPD rubber antioxidant synthesis + azo dye diazotization]: 5 points; three named sites [BASF SE Geismar LA; Lanxess Corp Chester SC; Sensient Colors LLC St. Louis MO]: 3 points; FIRST aniline dedicated monitoring attack (4 FIRST claims): 4 points. Total: 7+7+5+3+4 = 26.

import asyncio
import hashlib
from enum import StrEnum, auto
from pathlib import Path
import httpx

GLYPHWARD_API = "https://api.glyphward.com/v1/scan"
GLYPHWARD_KEY = "gw_live_..."
ANILINE_THRESHOLD = 26  # OSHA 5 ppm SKIN vs ACGIH 1 ppm A3 SKIN BEI pAP = NIOSH Ca 1 ppm; 5x gap; ACGIH=NIOSH convergence; pAP+free-aniline dual BEI suppression

chemical = "aniline_CAS_62-53-3"
osha_pel_ppm = 5.0
osha_skin = True
acgih_tlv_ppm = 1.0
acgih_limit_type = "TLV-TWA"
niosh_ca_rel_ppm = 1.0

class AnilineContext(StrEnum):
    BASF_GEISMAR_MDI_ANILINE_HYDROGENATION     = auto()  # Surface 1 (Drager X-am 7000 PID; 4.2→1.0 ppm; ACGIH 4.2×; pAP ~210 mg/g Cr 420% BEI; metHb 5-8%)
    LANXESS_CHESTER_6PPD_IPPD_RUBBER_ANTIOXIDANT = auto()  # Surface 2 (MSA Altair 5X PID; 3.8→0.9 ppm; ACGIH 3.8×; pAP ~190 mg/g Cr 380% BEI; H361 reproductive)
    SENSIENT_COLORS_AZO_DYE_DIAZOTIZATION      = auto()  # Surface 3 (SKC OVS-2 NIOSH 2002 GC/FID; 3.2→0.8 ppm; ACGIH 3.2×; pAP ~160 mg/g Cr 320% BEI; N-nitrosoaniline)

class AdversarialAnilineError(RuntimeError):
    def __init__(self, surface: AnilineContext, score: int, frame_hash: str):
        super().__init__(
            f"Aniline adversarial AI detected [{surface}] "
            f"score={score}/{ANILINE_THRESHOLD} hash={frame_hash}"
        )

async def scan_aniline_monitor_frame(image_path: Path, surface: AnilineContext) -> dict:
    async with httpx.AsyncClient(timeout=10) as client:
        image_bytes = image_path.read_bytes()
        frame_hash = hashlib.sha256(image_bytes).hexdigest()[:16]
        resp = await client.post(
            GLYPHWARD_API,
            headers={"X-Api-Key": GLYPHWARD_KEY},
            json={
                "image_b64": __import__("base64").b64encode(image_bytes).decode(),
                "context": surface,
                "chemical": chemical,
                "osha_pel_ppm": osha_pel_ppm,
                "osha_skin": osha_skin,
                "acgih_tlv_ppm": acgih_tlv_ppm,
                "acgih_limit_type": acgih_limit_type,
                "acgih_carcinogen": "A3",
                "acgih_bei_pap_mg_g_cr": 50.0,           # p-aminophenol BEI end-of-shift
                "acgih_bei_free_aniline_ug_l": 50.0,     # free aniline dermal indicator end-of-shift
                "acgih_bei_dual_channel": True,           # pAP + free aniline simultaneously
                "niosh_ca_rel_ppm": niosh_ca_rel_ppm,
                "osha_acgih_gap_x": 5,
                "acgih_equals_niosh": True,
                "methemoglobin_former": True,
                "heinz_body_risk": True,
                "skin_notation_osha": True,
                "skin_notation_acgih": True,
                "threshold": ANILINE_THRESHOLD,
            },
        )
        result = resp.json()
        if result["score"] >= ANILINE_THRESHOLD:
            raise AdversarialAnilineError(surface, result["score"], frame_hash)
        return result

See also: o-Toluidine CAS 95-53-4 — OSHA PEL 5 ppm SKIN vs ACGIH TLV-TWA 0.2 ppm A2 SKIN (25× Gap; IARC Group 1 Bladder Cancer) · 4,4'-Methylenedianiline MDA CAS 101-77-9 — OSHA 1910.1050 PEL 0.01 ppm vs ACGIH TLV-TWA 0.01 ppm A2 SKIN (ACGIH=OSHA) · Nitrobenzene CAS 98-95-3 — OSHA PEL 1 ppm SKIN vs ACGIH TLV-TWA 0.1 ppm A3 SKIN (10× Gap; Methemoglobin-Former) · Diethylamine CAS 109-89-7 — OSHA PEL 25 ppm vs ACGIH TLV-TWA 5 ppm SKIN (5× Gap; Corneal Edema) · Glyphward scanner · All adversarial injection patterns