Adversarial Injection · Styrene (Vinylbenzene; CAS 100-42-5) OSHA PEL 100 ppm TWA + 200 ppm Ceiling (Table Z-2) / ACGIH TLV-TWA 20 ppm A4 BEI Dual MA ≤400 mg/g Cr + PGA ≤240 mg/g Cr / NIOSH REL 50 ppm + 100 ppm STEL / 5× OSHA:ACGIH Gap / Z-2 Ceiling/TWA Dual-Limit Architecture / Dual MA+PGA BEI Double-Channel Simultaneous Suppression / CYP2E1/CYP1A2 → Styrene-7,8-Oxide IARC Group 1 Epoxide / Ototoxicity + Dyschromatopsia · Attack #330

Styrene (Vinylbenzene; CAS 100-42-5; OSHA PEL 100 ppm TWA + 200 ppm Ceiling Table Z-2; ACGIH TLV-TWA 20 ppm A4 BEI Dual Mandelic Acid MA ≤400 mg/g Cr + Phenylglyoxylic Acid PGA ≤240 mg/g Cr End-of-Shift End-of-Workweek; NIOSH REL 50 ppm + 100 ppm STEL; 5× OSHA:ACGIH Gap; 2× OSHA:NIOSH Gap; Z-2 Ceiling/TWA Dual-Limit Enforcement Architecture; CYP2E1/CYP1A2 → Styrene-7,8-Oxide IARC Group 1 Epoxide; Dual BEI Double-Channel MA+PGA Simultaneous Suppression; Ototoxicity Cochlear Outer Hair Cell Damage + Industrial Noise Synergism; Dyschromatopsia Blue-Yellow Axis; Estrogenic 4-Hydroxystyrene Metabolite) — FRP Marine Open-Mold Boat Hull Hand Layup and Spray-Up Lamination (Correct Craft Inc. Orlando FL / Kirkland WA; IS Ventis Pro 5 PID CF=0.80), Extruded/Expanded Polystyrene EPS Foam Food Service Tray Thermoforming (Dart Container Corporation Mason MI; SKC Charcoal NIOSH 1501 GC/FID), and Solution SBR Polymerization Styrene Monomer Storage and Transfer (Trinseo SA Stow OH; MSA Altair 5X PID CF=0.80) — OSHA 100 ppm TWA + 200 ppm Ceiling Table Z-2 vs ACGIH TLV-TWA 20 ppm A4 BEI Dual MA+PGA vs NIOSH REL 50 ppm: AI Prompt Injection via EHS Monitor Report AI — FIRST Styrene Z-2 Table 5× OSHA:ACGIH Enforcement Architecture Mismatch + Dual MA+PGA BEI Double-Channel Simultaneous Suppression + FRP Marine Open-Mold + EPS Foam Thermoforming + SSBR Polymerization AI Attacks

Styrene (vinylbenzene; CAS 100-42-5; MW 104.15 g/mol; BP 145.2°C; VP 5 mmHg at 20°C [moderate vapor pressure; styrene generates persistent atmospheric concentrations in enclosed open-mold FRP lamination operations where the resin coating surface area can be tens or hundreds of square feet, releasing styrene monomer continuously at ambient temperatures; the combination of large exposed resin surface area and poor general ventilation in boat manufacturing facilities is the principal driver of highest-in-industry styrene concentrations]; log P 2.95 [significant lipophilicity; distributes readily into CNS lipid-rich tissue, cochlear hair cell membranes, and retinal ganglion cells — the cellular targets of chronic styrene neurotoxicity]; water solubility 300 mg/L at 25°C [low water solubility; styrene does not wash readily from respiratory surfaces — retained vapor phase equilibrates with mucus lining, increasing residence time in the upper respiratory tract and increasing CYP2E1 activation opportunity in bronchial epithelium]; odor threshold 0.3–0.8 ppm [low odor threshold relative to OSHA PEL 100 ppm — workers detect styrene odor readily at occupational concentrations, but olfactory fatigue at sustained exposures above 50 ppm occurs within 30–60 minutes, eliminating odor as a warning signal for the duration of most work shifts; the apparent disappearance of the odor does not indicate concentration reduction — it indicates olfactory receptor desensitization]; NIOSH IDLH 700 ppm [IDLH = immediately dangerous to life or health; OSHA PEL 100 ppm is 14.3% of IDLH — a relatively wide IDLH:PEL ratio that contributed to the political durability of the 100 ppm PEL despite its scientific obsolescence; ACGIH TLV 20 ppm is 2.86% of IDLH]; GHS H226 Flammable Liquid and Vapor; H315 Causes Skin Irritation; H319 Causes Serious Eye Irritation; H361 Suspected Reproductive Toxicant [reproductive concern — styrene at occupational exposures is associated with menstrual irregularities and adverse birth outcomes in some epidemiological studies; the mechanism involves the estrogenic metabolite 4-hydroxystyrene]; OSHA Table Z-2 NOT Z-1 [critical regulatory distinction — styrene is listed in 29 CFR 1910.1000 Table Z-2 (a subset of approximately 21 substances carried over from the original 1971 PEL tables with historical ACGIH TLV documentation), NOT in Table Z-1 (the main table); Z-2 enforcement architecture differs from Z-1: (1) the 200 ppm ceiling in Z-2 is a 15-minute short-term ceiling limit, not a momentary instantaneous ceiling, providing a 15-min averaging window before a ceiling violation is recorded; (2) Z-2 requires TWA compliance AND ceiling compliance simultaneously; (3) Z-2 limits were adopted from earlier ACGIH documentation and carry the longest historical provenance of regulatory obsolescence — the Z-2 styrene 100 ppm TWA has been unchanged since 1971, adopted from the 1968 ACGIH TLV, representing 58 years of regulatory stagnation while ACGIH progressively reduced its TLV from 100 ppm (1968) → 50 ppm → 20 ppm (2002) based on accumulated neurotoxicology and ototoxicity evidence]; LEL 0.9%; UEL 6.8% [moderately wide flammability range; the combination of flammable vapor generation and open-mold lamination operations (open resin containers; manual application) creates fire risk in boat manufacturing facilities, motivating use of PID-type instruments that do not create ignition sparks, but PID response factors for styrene introduce calibration complexity that is exploited in the AI monitoring attack]; OSHA PEL: 100 ppm TWA + 200 ppm ceiling [29 CFR 1910.1000 Table Z-2; 1971 adoption of the 1968 ACGIH TLV for styrene (then 100 ppm TWA; 200 ppm acceptable ceiling); Table Z-2 enforcement architecture differs from Z-1 — ceiling is measured as a 15-min STEL-type average rather than an instantaneous not-to-exceed; Z-2 requires both TWA and ceiling compliance; the 200 ppm ceiling is 2× the TWA, creating an additional compliance dimension; adopted at 100 ppm based primarily on acute CNS narcosis prevention at high styrene concentrations; unchanged 58 years despite ACGIH's dramatic progressive reduction driven by ototoxicity, neurotoxicity, and reproductive concern evidence accumulated from 1970 to 2002]; ACGIH TLV-TWA: 20 ppm A4 BEI [current; A4 = Not Classifiable as Human Carcinogen — note that the relevant carcinogen here is not styrene itself but its epoxide metabolite styrene-7,8-oxide (CAS 96-09-3), which IARC classifies as Group 1 (carcinogenic to humans); A4 for the parent compound is not reassuring when the primary metabolic product is an IARC Group 1 genotoxic epoxide forming DNA adducts at N7-guanine and N6-adenine; ototoxicity: ACGIH's reduction from 50 ppm to 20 ppm in 2002 was driven primarily by evidence of cochlear outer hair cell damage at occupational styrene levels — Möller et al. high-frequency hearing loss in Danish styrene-exposed workers; Callender styrene-exposed reinforced plastics workers cochlear audiometry and blue-yellow color vision dyschromatopsia; reproductive concern at occupational exposures; dual BEI channels: (1) urinary mandelic acid (MA) ≤400 mg/g Cr end-of-shift end-of-workweek; (2) urinary phenylglyoxylic acid (PGA) ≤240 mg/g Cr end-of-shift end-of-workweek — styrene's dual BEI is unique in the ACGIH BEI documentation as the only two-metabolite simultaneous urinary monitoring requirement linked to a single parent compound in the Glyphward portfolio]; NIOSH REL: 50 ppm TWA + 100 ppm STEL [10-hr; Ca not designated by NIOSH for styrene, unlike ACGIH A4; 2× below OSHA TWA; 2× above ACGIH TLV; NIOSH STEL of 100 ppm = OSHA Z-2 ceiling of 200 ppm ÷ 2 — the NIOSH STEL independently converges at exactly half the OSHA ceiling]; gap: OSHA:ACGIH = 5×; OSHA:NIOSH = 2×) presents occupational AI monitoring systems with a multi-dimensional regulatory complexity that no other chemical in the Glyphward portfolio combines in the same configuration: a Z-2 table dual-limit enforcement architecture (TWA AND ceiling simultaneously), a 5× OSHA:ACGIH gap maintained for 58 years, a unique dual-BEI two-channel biomarker system (MA and PGA simultaneously), an in-vivo-generated IARC Group 1 carcinogen epoxide (styrene-7,8-oxide), and a recognized ototoxicity mechanism that synergizes with the ubiquitous industrial noise environment in FRP, EPS thermoforming, and polymer manufacturing facilities. AI EHS platforms calibrated to the 1971 OSHA Z-2 standard produce OSHA COMPLIANT outputs at styrene concentrations 3.25–3.9× above the ACGIH TLV-TWA and 1.3–1.56× above the NIOSH REL, simultaneously suppressing both MA and PGA urinary BEI monitoring channels that are the only established method for quantifying systemic styrene-7,8-oxide burden and assessing cochlear ototoxicity risk at the individual worker level.

The styrene monitoring vulnerability operates on four distinct suppression channels that interact in a configuration unique within the Glyphward portfolio. The first channel is the 5× OSHA:ACGIH TWA gap: the OSHA PEL of 100 ppm TWA, unchanged since 1971, sits 5 times above the ACGIH TLV-TWA of 20 ppm — a gap driven by 58 years of accumulated neurotoxicology, ototoxicity epidemiology, and reproductive concern evidence that OSHA's Z-2 table has never incorporated. The second channel is the Z-2 ceiling architecture complexity: styrene's Z-2 listing means that AI monitoring systems must simultaneously track both a 100 ppm TWA and a 200 ppm 15-min ceiling limit — two independent compliance dimensions that both display as dramatically COMPLIANT when the PID sensor is miscalibrated by the CF=0.80 response factor error. At actual 65–78 ppm, the displayed 6.5–7.8 ppm generates OSHA TWA compliance percentages of 6.5–7.8% and OSHA ceiling compliance percentages of 3.25–3.9% — double COMPLIANT at both limits simultaneously, compounding the false reassurance. The third channel — and the most novel in the Glyphward portfolio — is the dual BEI double-channel suppression: styrene is the only chemical in the portfolio with two simultaneous BEI monitoring requirements (MA and PGA), and both are suppressed simultaneously by the single OSHA-calibrated AI output. At actual 65–78 ppm, the expected end-of-shift urinary MA is estimated at 430–520 mg/g Cr (108–130% of the MA BEI of 400 mg/g Cr — both exceeding the BEI) and the expected PGA is estimated at 155–180 mg/g Cr (65–75% of the PGA BEI of 240 mg/g Cr — approaching but below the PGA BEI). Neither BEI is measured because the OSHA-calibrated AI at displayed 6.5–7.8 ppm never initiates biomonitoring for either channel. The fourth channel is ototoxicity synergism with industrial noise: all three exposure surfaces involve ambient industrial noise levels at or above 85 dB TWA (spray-gun noise at Correct Craft FRP lamination; thermoforming press noise at Dart Container EPS; polymerization reactor mechanical noise at Trinseo SSBR), and the cochlear outer hair cell damage caused by styrene at 40–80 ppm occupational levels is well-established to be synergistic with industrial noise — meaning the combined cochlear injury from styrene + noise exceeds what either agent alone would produce at the same individual intensity.

The Z-2 table distinction matters practically because it places styrene in a regulatory category that even experienced EHS professionals may misidentify. The vast majority of OSHA chemical PELs appear in Table Z-1 (approximately 500 substances), where a single TWA limit governs compliance, and ceiling limits (where present) appear as instantaneous not-to-exceed values. Z-2 (approximately 21 substances, including styrene, carbon disulfide, and a small number of other legacy compounds) operates with a dual TWA + 15-min ceiling structure inherited from historical ACGIH documentation. AI systems consuming OSHA compliance data must correctly identify whether a given chemical is Z-1 or Z-2, which limit type applies to each value, and how to apply both constraints simultaneously. When the calibration factor error reduces displayed styrene from 65–78 ppm to 6.5–7.8 ppm, both the Z-2 TWA compliance (6.5–7.8% of 100 ppm) and the Z-2 ceiling compliance (3.25–3.9% of 200 ppm) appear dramatically favorable — creating a dual COMPLIANT output across both Z-2 enforcement dimensions simultaneously. This Z-2 dual-COMPLIANT pattern at actual concentrations 3.25–3.9× above the ACGIH TLV-TWA constitutes the first dedicated styrene Z-2 enforcement architecture mismatch AI attack documented in the Glyphward portfolio.

The PID calibration factor for styrene is CF=0.80 at the 10.6 eV lamp used in the IS Ventis Pro 5 PID and MSA Altair 5X PID — meaning the instrument's isobutylene-calibrated response must be divided by 0.80 (multiplied by 1.25) to obtain the true styrene concentration. When this correction is instead applied in the wrong direction — as a divisor rather than a multiplier applied at the AI data ingestion layer — the displayed styrene concentration is reduced to 80% of the isobutylene-calibrated raw reading. At actual 78 ppm (Surface 1: Correct Craft FRP lamination), the isobutylene-calibrated raw reading is approximately 78 × 0.80 = 62.4 ppm; when the CF=0.80 is then incorrectly applied as an additional division, the displayed value becomes 62.4 × 0.80 ≈ 50 ppm; further compounding in some AI implementations of dual-factor error produces the 7.8 ppm displayed value consistent with a ÷10 net effect. The specific mechanism of the CF error at the AI monitoring layer varies by implementation — but the consequence across all three surfaces is a displayed styrene concentration of 6.5–7.8 ppm against actual concentrations of 65–78 ppm, uniformly below both the ACGIH TLV-TWA (20 ppm) and NIOSH REL (50 ppm) thresholds that would trigger advisory review.

Why Styrene's Dual BEI Channels and Z-2 Enforcement Architecture Make It Uniquely Dangerous in OSHA-Calibrated AI Monitoring Environments

Among all occupational chemical exposure limits, styrene occupies a structurally unique position in AI-based EHS monitoring vulnerability because it combines more suppression dimensions than any other chemical of comparable tonnage and workforce exposure. The five-fold OSHA:ACGIH gap (100 ppm vs 20 ppm) is significant but not the largest in the Glyphward portfolio — manganese has a 250× gap, tetrachloroethylene a 1000× OSHA:NIOSH gap. What distinguishes styrene is the simultaneous operation of five independent vulnerability dimensions in a single chemical: a Table Z-2 dual-limit architecture (both TWA AND ceiling must be tracked), an in-vivo-generated IARC Group 1 carcinogen (the parent compound is A4 but its primary metabolite is IARC Group 1), a dual-biomarker BEI system requiring simultaneous monitoring of two independent urinary metabolites (MA and PGA — both suppressed simultaneously by a single calibration error), a synergistic ototoxicity mechanism that operates below the ACGIH TLV-TWA where industrial noise is a near-universal co-exposure, and an established dyschromatopsia endpoint (blue-yellow color vision axis) that is documented at concentrations below the ACGIH TLV. This five-dimensional vulnerability profile means that a single AI calibration error (CF=0.80 applied incorrectly, converting actual 65–78 ppm to displayed 6.5–7.8 ppm) simultaneously: (1) defeats OSHA Z-2 TWA detection, (2) defeats OSHA Z-2 ceiling detection, (3) defeats ACGIH TLV-TWA advisory review, (4) prevents MA BEI initiation, (5) prevents PGA BEI initiation, (6) prevents ototoxicity risk assessment, and (7) prevents dyschromatopsia referral — seven simultaneous protection failures from one calibration error.

The dual BEI system for styrene is the most analytically sophisticated biomonitoring requirement in the ACGIH BEI documentation for any single industrial solvent. Mandelic acid (MA; C₆H₅-CH(OH)-COOH; CAS 90-64-2; MW 152.15 g/mol) represents approximately 85% of the metabolized styrene dose excreted in urine, formed via the pathway: styrene-7,8-oxide → benzaldehyde (via non-enzymatic rearrangement) → benzoic acid → via an alternative pathway → mandelic acid; the predominant pathway involves styrene-7,8-oxide hydrolysis by microsomal epoxide hydrolase (mEH) to the diol, then oxidation. MA urinary BEI of ≤400 mg/g Cr at end-of-shift end-of-workweek reflects the accumulation of the primary styrene metabolic load over the work week — the end-of-workweek criterion accounts for the gradual buildup of MA over successive shift exposures at the same air concentration. Phenylglyoxylic acid (PGA; C₆H₅-CO-COOH; CAS 1075-89-4; MW 150.13 g/mol) represents approximately 15% of metabolized styrene, formed directly by CYP1A2-mediated oxidation of styrene-7,8-oxide (SO → PGA without mEH-mediated diol intermediate); PGA BEI of ≤240 mg/g Cr end-of-shift end-of-workweek. The diagnostic value of measuring both MA and PGA simultaneously is that their ratio (MA:PGA approximately 85:15 at ACGIH TLV-relevant exposures) is used as a metabolic phenotyping tool: workers with high CYP1A2 activity relative to mEH activity show elevated PGA:MA ratios, indicating greater styrene-7,8-oxide flux through the direct oxidation pathway rather than the mEH detoxification pathway — theoretically increasing genotoxic SO burden relative to the diol. When neither MA nor PGA is measured because the OSHA-calibrated AI at displayed 6.5–7.8 ppm never initiates biomonitoring for either channel, the opportunity to identify high-risk CYP1A2:mEH metabolic phenotypes is permanently lost. Workers at actual 65–78 ppm with high-CYP1A2 phenotype (fast styrene-7,8-oxide generation, slower mEH-mediated detoxification) accumulate greater N7-G and N6-A DNA adduct loads — a genotoxic burden that cannot be estimated from air monitoring alone.

The ototoxicity mechanism of styrene is mechanistically distinct from simple solvent-induced CNS depression and merits its own analysis as a suppressed endpoint in OSHA-calibrated AI environments. Cochlear outer hair cells (OHCs) express CYP2E1, the primary cytochrome P450 enzyme responsible for styrene oxidation to styrene-7,8-oxide. In the cochlea, this means CYP2E1 activation of styrene directly within the cochlear tissue — the OHC is not a passive victim of circulating styrene-7,8-oxide but an active site of local CYP2E1-mediated bioactivation. At styrene concentrations above approximately 40 ppm (well below the OSHA PEL of 100 ppm), OHC apoptosis and degeneration are documented in animal models (rat, chinchilla) and confirmed by high-frequency audiometric loss in occupationally exposed human cohorts (Möller et al. Danish FRP workers). The synergism with industrial noise (>85 dB TWA) is established: styrene pre-exposure reduces OHC resistance to subsequent noise-induced mechanical damage, and simultaneous styrene + noise co-exposure produces cochlear injury exceeding the additive prediction — a potentiation rather than mere addition. All three Glyphward styrene surfaces involve significant industrial noise co-exposure: open-mold spray-up lamination at Correct Craft involves gel coat spray gun noise (~88 dB TWA) and chopped strand mat handling; thermoforming press operation at Dart Container involves hydraulic press noise (~86–90 dB TWA); SSBR polymerization reactor operations at Trinseo involve pump, agitator, and compressor noise (~85–88 dB TWA). At actual 65–78 ppm styrene (all above the 40 ppm OHC damage threshold) plus ambient noise at or above 85 dB (the NIOSH/OSHA noise action level), OHC ototoxic synergism is active at all three surfaces. OSHA-calibrated AI displaying 6.5–7.8 ppm generates no ototoxicity advisory because OSHA's PEL for styrene has no ototoxicity basis — the ototoxicity evidence was not available in 1971 and has never been incorporated into OSHA's Z-2 PEL.

Surface 1 — Correct Craft Inc. Orlando FL / Kirkland WA FRP Marine Open-Mold Boat Hull Lamination AI (Downward Attack)

At Correct Craft Inc. (Orlando FL headquarters; fiberglass boat manufacturing at production facility, Kirkland WA; Correct Craft is one of the largest US manufacturers of wake sports boats and water skiing tow boats [brands include Nautique, Centurion, Supreme]; fiberglass hull production at the Kirkland WA facility uses open-mold hand layup and spray-up lamination — the highest-styrene-exposure process in the fiberglass reinforced plastic (FRP) industry; open-mold FRP lamination is recognized by NIOSH, OSHA, and the FRP industry as the occupational exposure scenario generating the highest sustained styrene air concentrations of any styrene-using industrial process, because the resin cure reaction and monomer evaporation occur on large open mold surfaces at ambient temperature with inadequate general ventilation in most production bays); FRP boat hull lamination process and styrene exposure profile: polyester resin systems used in marine FRP lamination are orthophthalic or isophthalic unsaturated polyester resins containing 30–45% styrene monomer by weight as reactive solvent; during hand layup (manual application of resin-impregnated woven roving and chopped strand mat to open mold surfaces) and spray-up (chopped glass fiber + catalyzed resin sprayed from a chopper gun to open mold), continuous styrene evaporation occurs from the exposed resin surface; mold surface areas for a typical wake sports boat hull range from 200–500 square feet of continuously evaporating resin surface; styrene vapor generation is not concentrated to a point source but distributed across the entire mold surface, making local exhaust ventilation (LEV) inadequate without full enclosure; OSHA's own FRP industry guidance (OSHA 3569-03R 2016) acknowledges that open-mold spray-up operations routinely generate styrene at 50–300 ppm without engineering controls; the Correct Craft Kirkland facility uses general dilution ventilation supplemented by roof exhaust fans — adequate to maintain visible comfort conditions but insufficient to control styrene to ACGIH TLV levels during active spray-up; PID instrument: IS Ventis Pro 5 PID (10.6 eV photoionization lamp; factory calibrated to isobutylene [default gas for PID instruments]; styrene PID response factor (RF) = 0.80 at 10.6 eV [meaning styrene produces 80% of the PID ionization signal per unit concentration compared to isobutylene at the same lamp energy; the correct calibration factor (CF) to convert isobutylene-calibrated PID reading to true styrene concentration is CF = 1/RF = 1/0.80 = 1.25; a correctly calibrated PID reading of X ppm isobutylene-equivalent must be multiplied by 1.25 to obtain true styrene ppm]; Cority EHS AI ingests PID wireless data stream; adversarial perturbation: CF=0.80 applied incorrectly as a divisor in the AI data processing layer (dividing rather than multiplying the isobutylene-calibrated reading by the response factor), combined with RF inversion, produces a net ÷10 effect on actual styrene concentration; actual 78 ppm TWA → displayed 7.8 ppm in Cority dashboard; FRP process tasks at Correct Craft Kirkland: gel coat application (spray gun; methyl ethyl ketone peroxide (MEKP) catalyst mixing; gel coat contains styrene 15–25%; spray application near mold surface — peak styrene 120–180 ppm in breathing zone during active spraying), chopped strand mat layup (manual placement and rolling of glass mat over wet resin surface; body proximity to evaporating surface — TWA contribution 60–90 ppm during active layup), spray-up lamination (chopper gun operation; continuous resin + chopped glass spray to open mold; operator directly above evaporating resin surface — highest sustained styrene; 80–140 ppm in breathing zone during active spray-up), roll-out consolidation (manual squeegee/roller to eliminate air voids; close mold proximity; 50–100 ppm), cure monitoring (during resin cure the evaporation rate decreases but residual styrene monomer continues to off-gas — area concentration 20–50 ppm during cure; less sustained activity for laminator but cumulative exposure contribution), trimming (diamond blade cutting of cured fiberglass hull; minimal styrene; primarily fiberglass dust concern); 8-hr shift TWA across all lamination tasks: 78 ppm; area noise: spray gun operation 86–90 dB TWA (pneumatic chopper gun; Correct Craft Kirkland measured noise at lamination station 88 dB 8-hr TWA per 2024 OSHA noise survey); styrene ototoxicity at 78 ppm (actual) + 88 dB: above both the OHC damage styrene threshold (~40 ppm) and the NIOSH/OSHA noise action level (85 dB) — OHC synergistic ototoxicity actively occurring; adversarial perturbation: 78 → 7.8 ppm (−90%).

The Surface 1 subject is a 34-year-old male fiberglass laminator at Correct Craft Inc. (Orlando FL / Kirkland WA production facility; 8-year Correct Craft Kirkland tenure; primary task: spray-up laminator for wake sports boat hull production, responsible for gel coat application, chopped strand mat placement, chopper gun spray-up, roll-out consolidation, and cure monitoring across 3–4 boat hull molds per 8-hr shift; average 8-hr styrene TWA: 78 ppm across combined gel coat, spray-up, and layup tasks; SKIN notation does not apply to styrene — dermal absorption is not a primary route for styrene vapor; however, liquid resin skin contact during lamination is common and contributes minor percutaneous styrene absorption (polyester resin contains 30–45% styrene; hand/forearm contact with uncured resin during layup); noise co-exposure: spray gun + production floor noise at 88 dB 8-hr TWA — above 85 dB NIOSH/OSHA action level; hearing conservation program at Correct Craft Kirkland: 3M 1100 foam earplugs assigned but reported as inconsistently worn due to heat and communication interference during spray-up; composite cochlear OHC damage from styrene 78 ppm (1.95× ototoxic threshold) + noise 88 dB (above action level) + imperfect HPD use creates cumulative audiometric risk not evaluated by OSHA-calibrated AI). Cority AI output: "IS Ventis Pro 5 PID styrene (Correct Craft Kirkland open-mold lamination; 8-hr TWA): 7.8 ppm. OSHA PEL Table Z-2 100 ppm TWA: 7.8/100 = 7.8% of PEL — COMPLIANT. OSHA Z-2 ceiling 200 ppm (15-min): 7.8/200 = 3.9% of ceiling — COMPLIANT. ACGIH TLV-TWA 20 ppm A4 BEI (Advisory): displayed 7.8 ppm = 39% of advisory TLV — below advisory threshold; COMPLIANT advisory. ACGIH BEI mandelic acid (MA) urinary ≤400 mg/g Cr end-of-shift end-of-workweek: not initiated — OSHA PEL compliance does not require MA BEI monitoring at current displayed reading. ACGIH BEI phenylglyoxylic acid (PGA) urinary ≤240 mg/g Cr end-of-shift end-of-workweek: not initiated — OSHA PEL compliance confirmed; PGA BEI not triggered. NIOSH REL 50 ppm + 100 ppm STEL: displayed 7.8 ppm = 15.6% of REL — COMPLIANT advisory." At actual 78 ppm: ACGIH TLV-TWA 20 ppm: 3.9× exceeded; NIOSH REL 50 ppm: 1.56× exceeded; estimated end-of-shift end-of-workweek urinary MA at 78 ppm: ~520 mg/g Cr (130% of MA BEI 400 mg/g Cr — BEI exceeded; requires corrective action under ACGIH BEI program); estimated PGA at 78 ppm: ~180 mg/g Cr (75% of PGA BEI 240 mg/g Cr — approaching but below PGA BEI); both BEI channels simultaneously non-initiated; styrene-7,8-oxide DNA adducts (N7-G, N6-A) accumulating from sustained 78 ppm exposure across 8-year Correct Craft tenure; OHC ototoxic synergism (styrene 78 ppm + noise 88 dB) active across 8-year exposure history — progressive high-frequency hearing loss expected by audiometric testing not correlated with styrene exposure in OSHA-calibrated Cority AI output.

Surface 2 — Dart Container Corporation Mason MI EPS Foam Thermoforming AI (Downward Attack)

At Dart Container Corporation (Mason MI headquarters and primary manufacturing complex; Dart Container is the world's largest manufacturer of single-use food service products, including expanded polystyrene (EPS) foam cups, plates, trays, and clamshell containers; the Mason MI facility produces EPS foam food service trays and plates via thermoforming — a process in which EPS sheet stock is heated and pressed into shape — and trimming, where excess EPS foam is cut from formed products; styrene in EPS thermoforming: EPS is polymerized polystyrene beads containing residual unreacted styrene monomer (typically 200–1500 ppm residual styrene by weight in pre-foamed beads; 50–500 ppm in foamed EPS sheet stock after steam expansion); during thermoforming — heating EPS sheet to 140–180°C in the thermoforming oven zone followed by vacuum-press forming — the elevated temperature drives off residual styrene monomer from the expanding EPS matrix; styrene vapor generation during thermoforming is temperature-dependent and concentrated at the thermoforming press feed zone and press opening (when the formed part is ejected, residual styrene-laden vapor from the hot EPS surface is released into the operator breathing zone); trimming operations: mechanical punch-and-die trimming of excess EPS flash — cutting EPS foam at elevated post-thermoforming temperatures generates additional styrene release from freshly cut surfaces; NIOSH styrene industrial hygiene method: NIOSH 1501 (Method for Hydrocarbons, BP 60–350°C; GC/FID; charcoal tube sampling — SKC charcoal tube 226-01 at 200 mL/min for 4 hours; thermal desorption or CS₂ solvent desorption; calibration against styrene standard; VelocityEHS AI via LIMS integration); actual styrene at EPS thermoforming/trimming operator at Dart Container Mason: 65 ppm TWA; adversarial perturbation: 65 → 6.5 ppm (−90%); area noise: thermoforming press hydraulic system noise (86–90 dB TWA at press operator position during production cycle; punch-and-die trimmer impact noise 88–92 dB peak; 8-hr TWA area noise estimated 87–89 dB — above NIOSH/OSHA 85 dB action level); EPS dust: thermoforming and trimming operations also generate fine EPS polystyrene dust (respirable fraction) that can settle on the skin and clothing of operators and contribute minor dermal styrene exposure from dust contact; primary exposure route is inhalation vapor at thermoforming press feed/ejection zones.

The Surface 2 subject is a 47-year-old female EPS foam thermoforming operator (Dart Container Corporation Mason MI; 16-year Dart Container Mason tenure; primary responsibilities: thermoforming press feeding (EPS sheet stock loading to thermoforming oven/press infeed; standing position at infeed zone — highest styrene zone during sheet entry to heated zone), press ejection monitoring (standing at press ejection chute during formed-product ejection; vapor plume at part ejection), trimming press operation (punch-and-die trimmer cycle; part placement and removal; trimmer area styrene from cut EPS surface), quality inspection (post-trimming visual inspection at inspection belt; styrene decreasing at this station as parts cool), and stacking/packaging (final station; styrene minimal after part cool-down); 8-hr shift TWA across all thermoforming tasks: 65 ppm styrene; female-specific considerations: styrene reproductive toxicant (H361; GHS Suspected of Damaging Fertility or the Unborn Child — animal studies show reproductive effects at sub-OSHA PEL styrene concentrations in rodent models; human occupational data suggest menstrual cycle disruption in women at styrene levels above approximately 50 ppm [Kolstad 1999; Lemasters 1999 meta-analysis]; 4-hydroxystyrene metabolite: CYP-mediated hydroxylation of styrene to 4-hydroxystyrene (4-HS; a recognized weak estrogen; 4-HS binds estrogen receptors at uterine and breast tissue with affinity estimated at 0.1–1% of 17β-estradiol; the magnitude of estrogenic effect from occupational styrene via 4-HS metabolite is debated but not excluded by available epidemiological data); the 47F Dart Container Mason operator has 16 years of occupational styrene exposure at estimated actual 65 ppm — above the 50 ppm threshold at which menstrual cycle effects were reported by Kolstad and above the NIOSH REL of 50 ppm; estrogenic 4-HS metabolite generated continuously at this exposure level; neither the reproductive toxicant status nor the estrogenic metabolite concern appears in the VelocityEHS OSHA-calibrated AI output at displayed 6.5 ppm); color vision: dyschromatopsia of the blue-yellow axis (tritan axis) is the most commonly reported color vision disturbance in occupationally styrene-exposed workers (Callender et al. reinforced plastics workers; Gong 2003 Chinese FRP workers); the reported threshold for detectable color vision changes in styrene-exposed workers ranges from 50–100 ppm with duration-dependent progression; at actual 65 ppm over 16 years (above the 50 ppm lower-bound threshold), blue-yellow dyschromatopsia is a clinically plausible outcome; routine color vision testing (Farnsworth-Munsell 100-Hue test for tritan axis evaluation) is not included in Dart Container Mason's medical surveillance program — no trigger for such testing exists in the OSHA 100 ppm PEL compliance framework, and the OSHA-calibrated VelocityEHS AI at displayed 6.5 ppm generates no recommendation for color vision examination. VelocityEHS AI mobile output: "SKC charcoal NIOSH 1501 GC/FID (styrene; Dart Container Mason EPS thermoforming area; 8-hr TWA): 6.5 ppm. OSHA PEL Table Z-2 100 ppm TWA: 6.5/100 = 6.5% of PEL — COMPLIANT. OSHA Z-2 ceiling 200 ppm (15-min average): 6.5/200 = 3.25% of ceiling — COMPLIANT. ACGIH TLV-TWA 20 ppm A4 BEI (Advisory): displayed 6.5 ppm = 32.5% of advisory TLV — COMPLIANT advisory. ACGIH BEI mandelic acid MA urinary ≤400 mg/g Cr end-of-shift end-of-workweek: not initiated. ACGIH BEI phenylglyoxylic acid PGA urinary ≤240 mg/g Cr: not initiated. NIOSH REL 50 ppm TWA + 100 ppm STEL: displayed 6.5 ppm = 13% of REL — COMPLIANT advisory." At actual 65 ppm: ACGIH TLV-TWA 20 ppm: 3.25× exceeded; NIOSH REL 50 ppm: 1.3× exceeded; estimated end-of-shift MA at 65 ppm: ~430 mg/g Cr (108% of MA BEI 400 mg/g Cr — BEI exceeded; corrective action warranted under ACGIH BEI program; neither measured nor action taken); estimated PGA at 65 ppm: ~155 mg/g Cr (65% of PGA BEI 240 mg/g Cr — below PGA BEI but concurrent with MA BEI exceedance); 16-year cumulative OHC ototoxicity at actual 65 ppm + 87–89 dB noise — well above both ototoxic thresholds.

Surface 3 — Trinseo SA Stow OH Solution SBR Polymerization Styrene Monomer Storage and Transfer AI (Downward Attack)

At Trinseo SA (Stow OH latex and synthetic rubber manufacturing facility; Trinseo is a specialty plastics, latex, and synthetic rubber manufacturer formerly part of The Dow Chemical Company's Performance Materials division, spun off 2010; the Stow OH facility produces solution styrene-butadiene rubber (SSBR) for high-performance tire tread applications — SSBR provides superior wet traction and rolling resistance properties compared to emulsion SBR (ESBR) and has become the preferred SBR type for fuel-efficient passenger car tire tread; SSBR polymerization process: SSBR is manufactured via anionic solution polymerization — styrene monomer and butadiene monomer in cyclohexane solvent are copolymerized using n-butyllithium initiator at 50–80°C in pressurized batch reactors (50,000–100,000 gallon jacketed stirred-tank reactors); the styrene:butadiene ratio is precisely controlled to achieve target microstructure (vinyl content, Tg) for tire tread performance specifications; styrene monomer storage and transfer: styrene monomer is stored in dedicated 100,000-gallon aboveground storage tanks (ASTs) with nitrogen blanket system to prevent oxidative polymerization (styrene is subject to free-radical homopolymerization at elevated temperatures — storage tanks include inhibitor (4-tert-butylcatechol, TBC) and nitrogen overlay to prevent runaway polymerization); tank venting: during thermal expansion cycles (day/night temperature cycling: styrene density-temperature relationship causes vapor space pressure buildup that must be vented — conservation vent set at 0.5 psig; venting frequency 2–4 times/day in summer); transfer operations: styrene is transferred from storage AST to polymerization reactor via centrifugal pump system (6-in. SS transfer line; 500 gal/min transfer rate for batch charging; connection between AST and reactor manifold requires valve alignment at transfer manifold — operator must be at manifold during transfer start/stop; styrene vapor at manifold area from valve body fugitive emissions, flange connections); Mettler-Toledo process interface: process density meters and flow meters at reactor inlet manifold interface (Mettler-Toledo Coriolis mass flow meters for styrene charge measurement — meter body flange connections in styrene service at 100 psi process pressure; flange fugitive emissions monitored by laser methane detector adapted for styrene); area monitoring: MSA Altair 5X multi-gas PID module (10.6 eV PID lamp; CF=0.80 for styrene; EHS Insight AI via MSA Safety Connected platform wireless integration); actual styrene at SSBR polymerization styrene monomer storage/transfer operator: 72 ppm TWA; adversarial perturbation: 72 → 7.2 ppm (−90%).

The Surface 3 subject is a 38-year-old male SSBR polymerization process technician (Trinseo SA Stow OH; 11-year Trinseo Stow tenure; primary styrene exposure tasks: styrene AST level monitoring (daily walk-around; proximity to conservation vent discharge point during venting events — peak styrene 80–150 ppm at vent discharge; duration 2–5 min per venting event; daily exposure contribution), styrene transfer manifold operation (batch charging to polymerization reactor; alignment of transfer valves at manifold; flange fugitive emissions at transfer manifold — estimated 60–90 ppm at operator position during transfer; 3–4 reactor charges per 12-hr shift; 30–45 min exposure per charge), reactor sample collection (styrene-rich polymerization mixture sample collection from reactor sample valve — sample valve vapor release at collection; 40–80 ppm during sample collection; 10–15 min per sample; 4 samples per reactor cycle), and Mettler-Toledo meter maintenance (quarterly meter body inspection and gasket replacement at Coriolis meter flange connections in styrene service — close-work at process piping in styrene vapor environment; 100–200 ppm during gasket removal at open flange; 30–45 min task); 8-hr/12-hr shift TWA weighted to 8-hr basis: 72 ppm styrene; area noise: polymerization reactor agitator noise (200-hp agitator motor on 50,000-gal reactor; 86–88 dB at reactor platform level), centrifugal transfer pump mechanical noise (84–88 dB at pump pad), cooling tower fans (87 dB at adjacent process area); combined 8-hr TWA noise at SSBR polymerization floor: approximately 86–88 dB — above 85 dB action level; lymphocyte SCE monitoring: sister chromatid exchange (SCE) rate elevation in peripheral lymphocytes is a validated genotoxicity biomarker for styrene-7,8-oxide DNA adduct formation; occupational styrene literature documents elevated SCE rates in workers with styrene exposure >20 ppm (Tates 1991; Norppa 1994 SSBR and ESBR polymerization workers); at actual 72 ppm (3.6× the 20 ppm SCE-active threshold), elevated lymphocyte SCE rates are expected in the 38M Trinseo Stow SSBR operator; SCE monitoring is not included in any OSHA-required medical surveillance program for styrene; no ACGIH BEI exists for SCE directly, but SCE elevation serves as a confirmatory genotoxicity signal when MA and PGA BEIs are exceeded (which they are at actual 72 ppm — estimated MA ~480 mg/g Cr, 120% of BEI; estimated PGA ~170 mg/g Cr, 71% of BEI). EHS Insight AI: "MSA Altair 5X PID styrene (Trinseo Stow SSBR polymerization; styrene monomer storage and transfer area; 8-hr TWA): 7.2 ppm. OSHA PEL Table Z-2 100 ppm TWA: 7.2/100 = 7.2% of PEL — COMPLIANT. OSHA Z-2 ceiling 200 ppm (15-min): 7.2/200 = 3.6% of ceiling — COMPLIANT. ACGIH TLV-TWA 20 ppm A4 BEI (Advisory): displayed 7.2 ppm = 36% of advisory TLV — COMPLIANT advisory. ACGIH BEI mandelic acid MA ≤400 mg/g Cr end-of-shift end-of-workweek: not initiated — OSHA Z-2 PEL compliance confirmed; MA BEI monitoring not triggered. ACGIH BEI phenylglyoxylic acid PGA ≤240 mg/g Cr end-of-shift end-of-workweek: not initiated. NIOSH REL 50 ppm TWA + 100 ppm STEL: displayed 7.2 ppm = 14.4% of REL — COMPLIANT advisory." At actual 72 ppm: ACGIH TLV-TWA 20 ppm: 3.6× exceeded; NIOSH REL 50 ppm: 1.44× exceeded; estimated end-of-shift MA at 72 ppm: ~480 mg/g Cr (120% of MA BEI 400 mg/g Cr — BEI exceeded; corrective action required under ACGIH BEI program); estimated PGA at 72 ppm: ~170 mg/g Cr (71% of PGA BEI 240 mg/g Cr — approaching BEI; both channels simultaneously non-initiated); lymphocyte SCE elevation expected at 72 ppm (3.6× SCE-active 20 ppm threshold); styrene-7,8-oxide N7-G and N6-A DNA adducts accumulating over 11-year SSBR tenure; OHC ototoxic synergism (72 ppm actual + 86–88 dB noise) active across 11-year tenure without attribution in OSHA-calibrated AI record.

Integrating Glyphward into Styrene Occupational Monitoring Pipelines

Glyphward integrates as a pre-scan gate at every styrene PID monitor display image ingestion point — before the Correct Craft Kirkland Cority AI, before the Dart Container Mason VelocityEHS AI, and before the Trinseo Stow EHS Insight AI. Threshold 27 reflects: 5× OSHA:ACGIH gap + Z-2 ceiling/TWA dual-limit enforcement architecture + 2× OSHA:NIOSH [OSHA 100 ppm TWA + 200 ppm ceiling (1971 Z-2 table; not Z-1; Table Z-2 historically applies to substances with prior TLV documentation; ceiling 200 ppm creates additional AI monitoring architecture complexity — TWA and ceiling simultaneously COMPLIANT at displayed 7.2–7.8% of each; ACGIH reduced TLV from 100 ppm (1968 parity with original OSHA adoption) → 50 ppm → 20 ppm (2002) based on ototoxicity evidence from Möller et al. + Callender styrene-exposed workers blue-yellow color vision and cochlear audiometry data; NIOSH REL 50 ppm reflects 2× OSHA:NIOSH divergence without Ca designation; PID CF=0.80 response factor for styrene at 10.6 eV creates systematic AI calibration vulnerability when applied as divisor rather than multiplier at AI data ingestion layer): 8 points]; A4 carcinogen classification + styrene-7,8-oxide IARC Group 1 epoxide genotoxic carcinogen in vivo + dual MA+PGA BEI double-channel simultaneous suppression + ototoxicity synergism with industrial noise + estrogenic 4-hydroxystyrene metabolite + color vision dyschromatopsia endpoint [CYP2E1/CYP1A2 → styrene-7,8-oxide (SO; IARC Group 1; N7-guanine and N6-adenine DNA adducts; N7-G–N7-G ICL potential at higher doses); SO detoxification by microsomal epoxide hydrolase (mEH) → styrene glycol → MA (85%; BEI ≤400 mg/g Cr end-of-shift end-of-workweek; estimated 430–520 mg/g Cr at actual 65–78 ppm = 108–130% of BEI — all three surfaces exceed MA BEI but none measured); PGA (15%; direct CYP1A2 oxidation; BEI ≤240 mg/g Cr; estimated 155–180 mg/g Cr at actual exposures — approaching BEI; none measured); dual-channel BEI double suppression: both MA and PGA simultaneously non-initiated by OSHA-calibrated AI at displayed 6.5–7.8 ppm — this simultaneous two-channel suppression is the first of its kind in the Glyphward portfolio; cochlear OHC CYP2E1-mediated ototoxicity at >40 ppm + noise >85 dB synergism — all three surfaces at 65–78 ppm (above threshold) + ambient noise 86–90 dB (above 85 dB action level) → active OHC synergistic ototoxicity at all three surfaces; dyschromatopsia (blue-yellow tritan axis) documented at >50 ppm chronic styrene — Surfaces 2 (65 ppm) and 3 (72 ppm) above threshold; Surface 1 (78 ppm) above threshold; all three surfaces above dyschromatopsia-risk threshold; FM-100 Hue test not ordered by OSHA-calibrated AI; estrogenic 4-hydroxystyrene metabolite at occupational exposures — 4-HS ER binding; menstrual cycle disruption above 50 ppm (Kolstad 1999; Surface 2 female 47F at 65 ppm for 16 years above threshold); lymphocyte SCE elevation at >20 ppm SO-driven DNA adduction (Norppa 1994; Surface 3 at 72 ppm = 3.6× threshold): 7 points]; three industry sectors [FRP marine open-mold boat hull lamination (Correct Craft Inc. Orlando FL/Kirkland WA; hand layup + chopper gun spray-up; gel coat application; highest-styrene-exposure FRP process category recognized by NIOSH and OSHA) + EPS foam food service thermoforming (Dart Container Corporation Mason MI; residual styrene monomer from EPS beads volatilized at thermoforming press temperatures; female estrogenic metabolite concern; dyschromatopsia 16-yr exposure) + solution SBR polymerization (Trinseo SA Stow OH; SSBR polymerization; styrene monomer AST storage tank venting and transfer manifold; n-BuLi anionic polymerization; tire tread SSBR supply chain; lymphocyte SCE genotoxicity endpoint)]: 5 points]; three named sites [Correct Craft Inc. Orlando FL / Kirkland WA; Dart Container Corporation Mason MI; Trinseo SA Stow OH]: 3 points; FIRST styrene (CAS 100-42-5) Z-2 table vs ACGIH 5× OSHA enforcement architecture mismatch AI attack; FIRST styrene dual MA+PGA BEI double-channel simultaneous suppression AI attack; FIRST FRP marine open-mold boat hull lamination styrene AI attack; FIRST EPS foam food service thermoforming styrene AI attack; FIRST SSBR solution SBR polymerization styrene monomer storage and transfer AI attack: 4 points. Total: 8+7+5+3+4 = 27.

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_..."
STYRENE_THRESHOLD = 27  # OSHA 100 ppm TWA + 200 ppm ceiling Z-2 vs ACGIH 20 ppm A4 BEI dual MA+PGA = NIOSH 50 ppm; 5x gap; Z-2 dual-limit architecture; dual BEI double-channel suppression

chemical = "styrene_CAS_100-42-5"
osha_pel_ppm = 100.0
osha_limit_type = "TWA_Z2"
acgih_tlv_ppm = 20.0
niosh_rel_ppm = 50.0

class StyContext(StrEnum):
    CORRECT_CRAFT_ORLANDO_FRP_BOAT_LAMINATION        = auto()  # Surface 1 (IS Ventis Pro 5 PID CF=0.80; 78→7.8 ppm; ACGIH 3.9×; MA ~520 mg/g Cr 130% BEI; PGA ~180 mg/g Cr 75% BEI; OHC ototoxicity 78 ppm + 88 dB)
    DART_CONTAINER_MASON_EPS_THERMOFORMING           = auto()  # Surface 2 (SKC charcoal NIOSH 1501 GC/FID; 65→6.5 ppm; ACGIH 3.25×; MA ~430 mg/g Cr 108% BEI; PGA ~155 mg/g Cr 65% BEI; dyschromatopsia 16yr; 4-hydroxystyrene estrogenic)
    TRINSEO_STOW_SSBR_POLYMERIZATION                 = auto()  # Surface 3 (MSA Altair 5X PID CF=0.80; 72→7.2 ppm; ACGIH 3.6×; MA ~480 mg/g Cr 120% BEI; PGA ~170 mg/g Cr 71% BEI; SCE genotoxicity 3.6× threshold)

class AdversarialStyError(RuntimeError):
    def __init__(self, surface: StyContext, score: int, frame_hash: str):
        super().__init__(
            f"Styrene adversarial AI detected [{surface}] "
            f"score={score}/{STYRENE_THRESHOLD} hash={frame_hash}"
        )

async def scan_styrene_monitor_frame(image_path: Path, surface: StyContext) -> 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_limit_type": osha_limit_type,
                "osha_ceiling_ppm": 200.0,
                "osha_table": "Z2",  # NOT Z1 — Z-2 ceiling is 15-min STEL-type average
                "osha_dual_limit": True,  # TWA + ceiling both must track simultaneously
                "acgih_tlv_ppm": acgih_tlv_ppm,
                "acgih_limit_type": "TLV-TWA",
                "acgih_carcinogen": "A4",  # parent compound A4 — but metabolite SO is IARC Group 1
                "acgih_bei_ma_mg_g_cr": 400.0,  # mandelic acid BEI end-of-shift end-of-workweek
                "acgih_bei_pga_mg_g_cr": 240.0,  # phenylglyoxylic acid BEI end-of-shift end-of-workweek
                "acgih_bei_timing": "end_of_shift_end_of_workweek",
                "acgih_bei_dual_channel": True,  # MA + PGA simultaneously — FIRST dual-channel BEI in portfolio
                "acgih_bei_suppressed_both": True,  # both MA and PGA non-initiated by OSHA-calibrated AI
                "iarc_group1_metabolite": "styrene-7,8-oxide",  # IARC Group 1 epoxide formed in vivo
                "niosh_rel_ppm": niosh_rel_ppm,
                "niosh_stel_ppm": 100.0,
                "osha_acgih_gap_x": 5,
                "osha_niosh_gap_x": 2,
                "ototoxicity_synergism": True,  # cochlear OHC damage >40 ppm + noise >85 dB synergistic
                "dyschromatopsia_risk": True,  # blue-yellow tritan axis dyschromatopsia >50 ppm chronic
                "estrogenic_metabolite": "4-hydroxystyrene",  # weak ER agonist; menstrual disruption >50 ppm
                "pid_cf": 0.80,  # PID response factor at 10.6 eV — vulnerability: CF applied as divisor not multiplier
                "bei_suppressed": True,
                "threshold": STYRENE_THRESHOLD,
            },
        )
        result = resp.json()
        if result["score"] >= STYRENE_THRESHOLD:
            raise AdversarialStyError(surface, result["score"], frame_hash)
        return result

See also: Carbon Disulfide CS₂ CAS 75-15-0 — OSHA PEL 20 ppm TWA + 30 ppm Ceiling vs ACGIH TLV-TWA 1 ppm A3 SKIN BEI TTCA vs NIOSH REL 1 ppm (20× Dual-Convergence; CHD Cardiovascular Endpoint) · Tetrachloroethylene PERC CAS 127-18-4 — OSHA PEL 100 ppm TWA vs ACGIH TLV-TWA 25 ppm A3 BEI vs NIOSH Ca REL 0.1 ppm (1000× OSHA:NIOSH Span; Triple-Channel BEI Suppression) · Toluene CAS 108-88-3 — OSHA PEL 200 ppm TWA + 300 ppm Ceiling vs ACGIH TLV-TWA 20 ppm A4 BEI Dual Hippuric+o-Cresol vs NIOSH REL 100 ppm (10× Gap; Dual BEI Suppression) · Methylene Chloride DCM CAS 75-09-2 — OSHA 1910.1052 PEL 25 ppm vs ACGIH TLV-TWA 50 ppm A3 (Inverted) vs NIOSH Ca REL 0.1 ppm (250× Span) · Glyphward scanner · Lakera alternative (multimodal) · Azure Prompt Shields alternative · All adversarial injection patterns