Post-Accident Evidence, “Obvious Risk” to Trained Users, and Proof-Heavy Risk–Utility: Fifth Circuit Affirms Summary Judgment in Premises- and Products-Defect Claims

Case: George v. SI Group (5th Cir. July 9, 2026) (per curiam; not designated for publication).
Core rule crystallized: (1) post-accident investigative materials do not, without more, establish a premises owner’s pre-accident knowledge of a dangerous condition; (2) manufacturers/design participants owe no duty to warn trained foreseeable users of commonly known/obvious operational risks; and (3) a Texas design-defect claim fails at summary judgment when the plaintiff’s risk–utility showing and alternative-design proof are conclusory or unsupported by evidence of feasibility, cost, and impact on utility.

I. Introduction

This appeal stems from a severe scalding injury suffered by James George, a Veolia employee, when a hot-water tanker trailer (filled with ~200-degree water) tipped over on SI Group’s premises and dumped water through the open window of a nearby vacuum truck where George sat waiting. George sued:

  • SI Group, Inc. (premises owner) for premises liability under Texas common law;
  • Bulk Solutions, L.L.C. (distributor) for products liability;
  • Brenner Tank Services, L.L.C. (partial designer) for products liability.

The immediate precipitating events mattered: after an air bag on the tractor blew, Veolia personnel disconnected the tractor and left the loaded tanker trailer on its landing gear on gravel, without placing mats under the landing gear and with the top hatch open. Within minutes, a landing-gear leg sank, the trailer tipped, and scalding water entered the vacuum truck cab. The district court granted summary judgment to all remaining defendants and excluded George’s expert (Benjamin Gibson) on SI Group’s knowledge. The Fifth Circuit affirmed across the board.

Procedural context mattered as well. In the prior appeal, George v. SI Grp., Inc., 36 F.4th 611 (5th Cir. 2022), the Fifth Circuit had remanded SI Group’s premises claim for Texas common-law analysis (rather than Chapter 95) and revived claims against certain product-side defendants. On remand, after further discovery and reassignment, the district court again disposed of the case on summary judgment—now affirmed.

II. Summary of the Opinion

A. Premises Liability (SI Group)

  • The court applied Clayton W. Williams, Jr., Inc. v. Olivo, 952 S.W.2d 523 (Tex. 1997), distinguishing two premises-defect settings for contractor employees.
  • Type 1 (premises condition not created by contractor’s work): George failed to raise a fact issue that SI Group had actual or constructive pre-accident knowledge of an unsafe condition. Post-accident reports did not show pre-accident knowledge; testimony suggesting a pre-accident “leak” was speculative and (as characterized) hearsay; and other testimony did not show deficient inspection or knowledge.
  • Type 2 (condition created by contractor’s work activity): George did not identify a relevant “condition created by the independent contractor’s work” that caused injury, and he did not show SI Group retained or actually exercised the required right of control over the activity.
  • Exclusion of George’s expert under Federal Rule of Evidence 702 was affirmed because the expert did not bring specialized expertise to the pre-accident knowledge question; he merely drew conclusions from record materials.

B. Products Liability (Bulk and Brenner)

  • Bulk: affirmance was automatic because George did not challenge on appeal the district court’s ruling that Bulk was entitled to non-manufacturing seller protection under Texas Chapter 82; the Fifth Circuit invoked appellate forfeiture under Rollins v. Home Depot USA, 8 F.4th 393 (5th Cir. 2021).
  • Brenner—failure to warn (marketing defect): no duty to warn because the risk of uncoupling a loaded trailer on gravel without stabilizing mats is commonly known/obvious to foreseeable users with special training, under Humble Sand & Gravel, Inc. v. Gomez, 146 S.W.3d 170 (Tex. 2004), and Sauder Custom Fabrication, Inc. v. Boyd, 967 S.W.2d 349 (Tex. 1998) (per curiam).
  • Brenner—design defect: even assuming disputes about causation, the Fifth Circuit affirmed on an alternative ground: George did not produce sufficient evidence that the trailer was defectively designed so as to be “unreasonably dangerous” under the Texas risk–utility framework described in Timpte Indus., Inc. v. Gish, 286 S.W.3d 306 (Tex. 2009). Conclusory assertions about “readily available” larger landing-gear feet and wider landing gear lacked evidence of feasibility, cost, safety, and effect on utility.

III. Analysis

A. Precedents Cited and How They Drove the Result

1. Summary-judgment framework and appellate standards

  • Guillory v. Domtar Indus. Inc., 95 F.3d 1320 (5th Cir. 1996): supplied the de novo review standard for summary judgment.
  • In re Deepwater Horizon, 48 F.4th 378 (5th Cir. 2022): reinforced that summary judgment is appropriate when “critical evidence” is too weak to support the essential fact.
  • Pierce v. Dep't of U.S. Air Force, 512 F.3d 184 (5th Cir. 2007): the court must view evidence in the light most favorable to the nonmovant.

These authorities framed the panel’s approach: it was not re-trying the case; it was asking whether George produced competent evidence capable of supporting required elements (knowledge, control, duty to warn, unreasonably dangerous design) beyond speculation.

2. Expert admissibility

  • In re Complaint of C.F. Bean L.L.C., 841 F.3d 365 (5th Cir. 2016), and Roman v. W. Mfg., Inc., 691 F.3d 686 (5th Cir. 2012): defined abuse-of-discretion review and the “wide latitude” district courts have in expert gatekeeping.
  • United States v. Wen Chyu Liu, 716 F.3d 159 (5th Cir. 2013): emphasized the “fit” requirement—specialized knowledge must be sufficiently related to the issues to assist the trier of fact.

Applying these, the panel upheld exclusion of Gibson because his “knowledge” did not materially exceed what a juror could infer from the same documents; his opinion on SI Group’s pre-accident knowledge was essentially argument, not expert methodology.

3. Texas premises-defect doctrine for contractor employees

  • Clayton W. Williams, Jr., Inc. v. Olivo, 952 S.W.2d 523 (Tex. 1997): supplied the organizing framework distinguishing (i) pre-existing/independent premises defects (duty hinges on owner’s knowledge) from (ii) defects created by contractor activity (duty hinges on retained/actual control related to the injury-causing condition/activity).
  • Sampson v. Univ. of Tex. at Aus., 500 S.W.3d 380 (Tex. 2016): cited to underscore that negligent activity claims are distinct from premises defects (and to correct the district court’s mislabeling), though the “right to control” concept overlaps.

How they influenced the result: Once the court framed George’s case as a premises-defect claim (not negligent activity), the burden centered on: (a) SI Group’s pre-accident knowledge for Type 1; and (b) the identification of a contractor-created defect plus SI Group’s retained/actual control for Type 2. The panel found George’s evidence lacking on both.

4. Texas product warnings and “obvious risk” to trained users

  • Am. Tobacco Co., Inc. v. Grinnell, 951 S.W.2d 420 (Tex. 1997): stated that duty to warn is a question of law and that failure to warn is a marketing defect.
  • Humble Sand & Gravel, Inc. v. Gomez, 146 S.W.3d 170 (Tex. 2004): no duty to warn of risks commonly known to foreseeable users; and when users have special training, no duty to warn of risks that should be obvious to them.
  • Sauder Custom Fabrication, Inc. v. Boyd, 967 S.W.2d 349 (Tex. 1998) (per curiam): an “obvious” risk to trained workers defeats duty to warn as a matter of law.

How they influenced the result: The panel treated commercial drivers as the foreseeable user class and relied on the presence of specialized training (commercial driver licensing and standard operating practices) to classify the risk—uncoupling a loaded trailer on unsupported gravel without mats—as obvious/common knowledge, eliminating a legal duty to add warnings.

5. Texas design defect: producing cause and risk–utility

  • Timpte Indus., Inc. v. Gish, 286 S.W.3d 306 (Tex. 2009): provided elements (defectively designed/unreasonably dangerous; safer alternative design; producing cause) and enumerated the risk–utility factors.
  • Ford Motor Co. v. Ledesma, 242 S.W.3d 32 (Tex. 2007): defined producing cause (substantial factor + but-for) and recognized multiple producing causes.
  • Goodner v. Hyundai Motor Co., Ltd., 650 F.3d 1034 (5th Cir. 2011): causation is typically for the jury, but can be resolved as a matter of law when evidence points strongly against causation.
  • Holtzclaw v. DSC Commc'ns Corp., 255 F.3d 254 (5th Cir. 2001): allowed affirmance on any ground supported by the record, even if different from the district court’s rationale.
  • Hernandez v. Tokai Corp., 2 S.W.3d 251 (Tex. 1999): invoked to support resolving unreasonably dangerous design under risk–utility as a matter of law where factors weigh decisively.

How they influenced the result: The panel flagged a potential analytical issue in the district court’s causation reasoning (because Ford Motor Co. v. Ledesma allows multiple producing causes), but it avoided reversal by applying Holtzclaw v. DSC Commc'ns Corp. and affirming on a cleaner basis: George’s inadequate proof that the design was unreasonably dangerous under Timpte Indus., Inc. v. Gish—especially the lack of evidence on feasibility/cost/utility impacts of the proposed alternative designs.

6. Appellate forfeiture

  • Rollins v. Home Depot USA, 8 F.4th 393 (5th Cir. 2021): failure to adequately brief an issue forfeits it.

This controlled the disposition for Bulk: because George did not challenge the Chapter 82 non-manufacturing seller ruling, the Fifth Circuit affirmed without reaching the merits.

B. Legal Reasoning: How the Court Reached Its Conclusions

1. Type 1 premises defect: knowledge must be pre-accident and evidence must be competent

Under Clayton W. Williams, Jr., Inc. v. Olivo, Type 1 liability depends on whether the premises owner knew or should have known of the dangerous condition. The Fifth Circuit treated George’s proof as failing for two recurring reasons:

  • Temporal mismatch: post-accident investigative reports may describe what was learned after the fact, but they do not themselves show what SI Group knew or should have known beforehand.
  • Speculation (and evidentiary weakness): testimony about overheard statements and assumptions about timing did not reliably establish pre-accident knowledge; the court characterized the inference chain as speculative rather than probative.

2. Rule 702 “fit”: experts cannot simply re-argue the record as “knowledge”

The court accepted the district court’s gatekeeping determination that Gibson’s opinions on SI Group’s knowledge were not based on a specialized method applied to specialized facts; instead, he “merely evaluates” the same evidence and “draws his own conclusions.” Under United States v. Wen Chyu Liu and Rule 702, that lack of “helpfulness” and specialized nexus justified exclusion.

3. Type 2 premises defect: identify the contractor-created condition and prove retained/actual control

For Type 2 defects, the “right to control” (or actual exercise of control) must relate to the injury-causing activity/condition. The opinion emphasizes two practical requirements:

  • Pleading/proof coherence: George did not clearly identify the contractor-created condition that caused the injury (as distinct from general unsafe conduct), undermining the Type 2 theory’s premise.
  • Control evidence must be concrete: the Veolia contract assigned safety precautions and equipment adequacy inspections to Veolia. A witness’s statement that SI Group could have required mats was not evidence that SI Group did exercise control “contrary to the contract’s terms,” as contemplated by Clayton W. Williams, Jr., Inc. v. Olivo.

4. Failure to warn: foreseeable users’ specialized training can eliminate duty

The panel’s warning analysis is a straightforward application of Am. Tobacco Co., Inc. v. Grinnell, Humble Sand & Gravel, Inc. v. Gomez, and Sauder Custom Fabrication, Inc. v. Boyd. The opinion treats commercial drivers of tanker trailers as a trained user group:

  • operation requires a commercial driver’s license (citing Texas’s CMV framework, including Tex. Transp. Code § 522.011(a));
  • training materials and workplace practice addressed the need to ensure the parking surface supports the trailer and to use mats when uncoupling a loaded trailer on non-concrete surfaces.

On those facts, the risk was legally “obvious” to foreseeable users, so Brenner had no duty to warn.

5. Design defect: the risk–utility test is evidentiary, not rhetorical

The Fifth Circuit’s most instructive product-liability move is its reliance on Timpte Indus., Inc. v. Gish risk–utility factors to affirm as a matter of law. Even though George pointed to two purported alternative designs (larger landing-gear feet; wider landing gear), the court found the showing incomplete:

  • no meaningful evidence that the injury risk “outweighed” the trailer’s utility given trained users’ ability to avoid the hazard (mats/level surface);
  • no supported evidence on whether alternatives would be unsafe, unreasonably expensive, or would impair utility / increase costs;
  • users’ anticipated awareness and ordinary consumer expectations both cut against labeling the design “unreasonably dangerous” when the hazard is avoidable by standard procedures.

The result underscores that risk–utility analysis often turns on concrete feasibility and cost proof—especially at summary judgment.

C. Impact: What This Opinion Signals for Future Litigation

  • Premises cases involving contractors: Plaintiffs should expect courts to demand pre-incident knowledge evidence tied to the defendant’s inspection, maintenance history, prior incidents, or contemporaneous complaints. Post-accident reports, standing alone, are unlikely to create a fact issue on knowledge.
  • Control-based liability theories: “Could have required” safety measures is not the same as “retained the right” or “actually exercised” control. Contracts allocating safety responsibility to the contractor will remain powerful unless contradicted by evidence of actual supervision over the specific activity that caused injury.
  • Warnings to trained users: The “special training/obvious risk” doctrine continues to be a robust duty-limiting principle in Texas products cases, especially for commercial/industrial equipment. Plaintiffs will need to reframe duty arguments around non-obvious hazards, hidden failure modes, or foreseeable misuse not covered by training.
  • Design defect proof burden: Merely naming an alternative design (bigger feet; wider stance) is not enough. Plaintiffs should anticipate the need for admissible engineering/economic evidence (feasibility, cost, performance tradeoffs, safety consequences) that maps directly onto Timpte Indus., Inc. v. Gish factors.
  • Appellate practice: The Bulk affirmance is a reminder that Chapter 82 defenses can become effectively unassailable on appeal if not directly briefed, per Rollins v. Home Depot USA.

Because the opinion is not designated for publication, its precedential force is limited under circuit rules, but its reasoning is likely to be persuasive to district courts confronting similar evidentiary gaps in premises and product cases.

IV. Complex Concepts Simplified

  • Type 1 vs. Type 2 premises defect (from Clayton W. Williams, Jr., Inc. v. Olivo):
    • Type 1: a dangerous condition that exists independently of the contractor’s work (e.g., an unsafe surface the owner maintained). Liability typically requires proof the owner knew/should have known of it.
    • Type 2: a dangerous condition created by the contractor’s own work activity. The owner usually owes no duty unless it retained (or actually exercised) control over the injury-causing activity/condition.
  • “Duty to warn” as a question of law: Courts decide whether a warning is legally required before a jury ever considers whether a warning was adequate. If the risk is obvious to trained foreseeable users, there is often no duty to warn.
  • Producing cause (Texas): A cause is “producing” if it is a substantial factor and the harm would not have occurred without it. There can be more than one producing cause (Ford Motor Co. v. Ledesma).
  • Risk–utility (design defect): This is a structured balancing test asking whether the design’s risks outweigh its benefits, considering feasibility and costs of safer designs, what users would expect, and whether warnings/procedures make the risk avoidable (Timpte Indus., Inc. v. Gish).
  • Rule 702 “fit” (experts): An expert must add specialized knowledge that helps the factfinder. If the expert simply reviews documents and reaches common-sense conclusions, courts may exclude the testimony as unhelpful.

V. Conclusion

The Fifth Circuit’s affirmance in George v. SI Group reinforces a practical evidentiary message across two domains.

  • For premises liability: pre-accident knowledge and concrete control evidence are essential; post-accident investigations and speculative testimony will not substitute.
  • For products liability: trained-user “obvious risk” doctrine can defeat warning claims as a matter of law, and design-defect plaintiffs must substantiate risk–utility and safer-alternative design elements with real feasibility and cost evidence.

In combination, the opinion functions as a roadmap for summary judgment: it is not enough to show a tragic outcome; the plaintiff must connect admissible proof to each doctrinal element—knowledge, duty, control, defect, and causation—under the governing Texas frameworks.