The Algorithmic Workplace: How AI Governance and Automated Management Expose Employers to Unfair Labor Practices and Systemic Discrimination Liability

by Jozy Allen, Associate Member, University of Cincinnati Law Review Vol. 95

I. Introduction

“Your scientists were so preoccupied with whether or not they could do it, they didn’t stop to think if they should.”¹

The rapid integration of Artificial Intelligence (“AI”) into today’s workplace constitutes a fundamental paradigm shift in the strategic architecture of the employer-employee relationship.² As organizations move toward algorithmic governance, the “black box” nature of these systems creates an epistemological gap that questions traditional legal doctrines of transparency, intent, and agency.³ The internal logic of deep-learning models, which is often unclear even to the people who built them, makes it difficult for employers to provide the “legitimate, non-discriminatory reason” required to rebut a prima facie case of unfair treatment under the McDonnell Douglas Corp. v. Green framework.⁴ This lack of clarity creates a profound liability vacuum: when an algorithm serves as a silent arbiter of a promotion, termination, or recruitment rejection, the organization may not be able to explain the decision-making process for judicial scrutiny.⁵

This Post proposes a strategic “human-in-the-loop” (“HITL”) governance model to minimize structural liabilities under the National Labor Relations Act (“NLRA”), Title VII, the Americans with Disabilities Act (“ADA”), and an array of state-level privacy and employment regulations.⁶ It delineates the increasing tension between the pursuit of technological efficiency and the preservation of established job protections. The Background section reviews statutory NLRA collective bargaining obligations under Section 8(d) and the evolving federal administrative landscape governing electronic monitoring. The Discussion section breaks down steps for organizations: how to write clear employee handbooks, set up required bargaining processes before using algorithmic monitoring, reduce legal risks tied to agency relationships and vendor agreements, and meet unique state rules for automated hiring tools. Finally, the Conclusion pulls these ideas together to offer employers concrete strategies they can put into action.

II. Background

A. The NLRA & Collective Bargaining in the Age of AI

The NLRA provides the primary regulatory scheme governing technological integration within unionized labor relations when it involves an employer’s collective bargaining obligations.⁷ For organizations with a unionized workforce, the introduction of AI is not a pure managerial prerogative; rather, it is a mandatory subject of bargaining under Section 8(d) whenever AI substantively alters “wages, hours, and other terms and conditions of employment.”⁸ Section 7 of the NLRA protects employees’ rights to form or join labor unions, to bargain collectively through self-elected representatives, and to take part in protected concerted activities (“PCA”).⁹ Courts’ clarifications in this area rest upon the distinction between the core decision to automate, often protected as a managerial right, and the mandatory duty to bargain over the effects of that automation on the bargaining unit.¹⁰

B. Good Faith Negotiation Requirements for Automation

When workplace automation strips away human jobs, the NLRA triggers a mandatory duty to bargain in good faith.¹¹ Section 8(d) requires employers and unions to meet at reasonable times and genuinely confer over terms and conditions of employment.¹² Neither side must concede or accept a proposal, but both must bring a sincere effort to reach common ground.¹³ The Supreme Court drew the line in the sand in First National Maintenance Corp. v. NLRB, creating a balancing test: while a company’s core, economically driven business decisions are protected managerial prerogatives, employers cannot simply walk away.¹⁴ They must still bargain in good faith over the effects that workforce displacement has on the impacted workers.¹⁵ A consistent stream of labor precedents reinforces that whenever technology alters job duties or redefines bargaining duties, negotiations precede rollout.¹⁶

Yet management is hardly powerless. As the Eighth Circuit emphasized in Omaha Typographical Union v. NLRB, employers can drive a hard bargain to push through crucial automation needed to stay competitive in shifting markets.¹⁷ So long as the employer negotiates the human impacts in good faith, holding a steady stance on key terms does not amount to unlawful surface bargaining.¹⁸

C. Legal Risks of AI Management Tools and Automated Monitoring

The duty to bargain under the NLRA extends beyond layoff decisions.¹⁹When an employer uses AI management tools,²⁰ it directly alters “terms and conditions of employment” governed by Section 8(d) by substantially altering daily working conditions and performance expectations.²¹ Unilaterally deploying automated monitoring or discipline without first bargaining to agreement or a genuine impasse creates immediate Section 8(a)(5) liability.²² Worse, if an employer relies on algorithmic metrics to discipline or discharge employees involved in PCA, the NLRB analyzes the retaliation claim under the Wright Line burden-shifting framework.²³ Once the General Counsel (”GC”) shows that an employee’s Section 7 activity motivated the adverse action, the employer can escape liability only by proving that it would have taken the exact same measure under neutral, non-discriminatory standards alone.²⁴

D. Shifting NLRB Guidance on AI Surveillance

The political pendulum at the NLRB complicates compliance, as administrative guidance shifts with every presidential transition.²⁵ Over recent years, NLRB GC guidance on electronic surveillance has swung dramatically.²⁶ From GC Peter Robb’s more restrictive approach to expansive readings of Section 7 rights under the first Trump administration,²⁷ to GC Jennifer Abruzzo’s Biden-era GC Memo 23-02, which proposed that intrusive automatic monitoring and management has a tendency to interfere with Section 7 rights,²⁸ to Acting GC William Cowen’s rescission of that memo through GC Memo 25-05 in the second Trump administration.²⁹ By pulling GC Memo 23-02 off the books, Abruzzo’s proposed framework was withdrawn, leaving the Board’s existing precedent and applicable jurisprudence to govern unless and until the Board adopts a different standard.³⁰

III. Discussion

A. Avoiding Implied Contracts: The Importance of Clear Handbook Disclaimers

The Discussion section breaks down steps for organizations: how to write clear employee handbooks, set up required bargaining processes before using algorithmic monitoring, reduce legal risks tied to agency relationships and vendor agreements, and meet unique state rules for automated hiring tools.

Employee handbooks can be strategic governance tools but, if improperly drafted, can become a liability that undermines at-will employment.³¹ As employers scramble to integrate AI usage policies, they face a hidden contractual trap: converting procedural promises about algorithmic review into enforceable commitments.³² Courts do not hesitate to treat handbook provisions as binding obligations when employees acknowledge and rely on granular handbook provisions, as shown in the landmark rulings in Quijano v. Atkins-Kroll, Inc. and D’Angelo v. Gardner.³³ Drafting a rigid, multi-tier human appeal process for algorithmically flagged infractions? The employer may unintentionally forfeit its at-will termination right.³⁴ Compounding the risk, Watson v. Idaho Falls warned that requiring employees to sign formal acknowledgments, though standard practice for proving notice, can actually strengthen the judicial inference that the handbook forms an enforceable agreement.³⁵

To avoid the danger of creating an implied contract through handbook policies, employers must put a clear, conspicuous disclaimer in their handbooks.³⁶ South Carolina law requires text to be underlined and capitalized on the first page of the handbook.³⁷ This mechanism keeps the at-will status while still giving necessary behavioral guidance.³⁸ By using this highly visible method, organizations across jurisdictions can ensure that their AI policies remain strictly advisory rather than legally enforceable contractual covenants.³⁹

B. Bargaining Before Algorithmic Monitoring: Ensuring NLRA Compliance

In workplaces with unions, adding AI tracking or automated performance standards to employee handbooks is not a routine administrative revision.⁴⁰ It is a change to mandatory bargaining subjects.⁴¹ Section 8(a)(5) makes clear that if an employer unilaterally alters terms and conditions⁴² without negotiating to agreement or good-faith impasse with a certified union, it is an unlawful refusal to bargain.⁴³ Consequently, employers must establish formal Section 8(d) bargaining channels before implementing algorithmic oversight.⁴⁴

The legal risk goes past Section 8(a)(5) obligations, as restricting employees from sharing AI-generated performance data or schedules exposes employers to Section 8(a)(1) liability by chilling PCA.⁴⁵ The NLRB treats employees sharing AI-generated schedules or performance statistics to resolve wage disputes or workplace grievances as PCA under Section 7, because workers are using employer-generated metrics to discuss and improve working conditions.⁴⁶ As a result, blanket bans on sharing algorithmic workplace data are presumptively unlawful under the NLRB’s Stericycle Inc. framework, which evaluates work rules from the perspective of an economically dependent employee.⁴⁷ While management attorneys often view this initial burden as setting a low threshold for workers, employers can defend a restrictive rule by proving it advances a legitimate, substantial business interest that cannot be achieved through a less restrictive policy.⁴⁸

C. Reducing Legal Risks in Automated Hiring: Human Oversight and Vendor Agreements

To mitigate vicarious liability under traditional agency principles, employers should structure vendor agreements to keep final hiring decisions in human hands rather than fully outsourcing evaluation to third-party AI software.⁴⁹ Courts are increasingly examining whether algorithm bias might be attributed directly to employers through their software vendors.⁵⁰ In Mobley v. Workday, Inc., a federal district court allowed claims against an HR software vendor to proceed, holding that an AI vendor may be independently liable as an “agent” under Title VII, the ADA, and the Age Discrimination in Employment Act (”ADEA”) when an employer delegates traditional screening and rejection functions to automated software.⁵¹

In contrast, the court in Saas v. Major, Lindsey & Africa LLC dismissed an algorithmic discrimination claim because the plaintiff relied on broad speculation rather than specific facts establishing a discriminatory selection process or measurable adverse impact.⁵² When read together, Mobley and Saas show that employers must back up hiring choices with legitimate, non-discriminatory documentation alongside contractually embedding HITL oversight to defeat speculative bias lawsuits.⁵³ Even so, the financial exposure associated with automated screening is significant, as shown by the Equal Employment Opportunity Commission’s (”EEOC”) $365,000 settlement in EEOC v. iTutorGroup, Inc. over software that automatically rejected older applicants based on hardcoded birth dates.⁵⁴

Beyond Title VII and the ADEA, automated screening tools can also create substantial exposure under the Americans with Disabilities Act.⁵⁵ The EEOC warns that AI screening violates the ADA if it disqualifies applicants with disabilities who could perform essential job functions with reasonable accommodations.⁵⁶ Often, automated tools that assess facial expressions, speech patterns, or personality in video interviews frequently penalize applicants who have neurological impairments.⁵⁷ To comply with the ADA, employers must tell applicants exactly what traits the AI is checking for, provide instructions for requesting reasonable accommodations before screening, and offer other methods to apply when disabled candidates are disadvantaged.⁵⁸

D. Meeting Jurisdiction-Specific Requirements: State and Local AI Hiring Laws

Tying all these exposure points together is a confusing patchwork of state and local regulations that forces multi-state employers into continuous auditing and localized compliance.⁵⁹ For example, Illinois requires advance notice, explanation for AI logic, and explicit candidate consent for AI video evaluations, while its Human Rights Act bars using zip codes as proxies for race.⁶⁰ New York City’s Local Law 144 forces employers to run annual bias audits and publish result summaries before using Automated Employment Decision-Making Tools.⁶¹ Lastly, Colorado places heavy affirmative duties on deployers of AI, imposing continuous risk protocols, impact assessments, adverse-action notices explaining rejection rationales, and granting candidates explicit rights to correct data and request appeals.⁶² Fair Credit Reporting Act class action risks and the legal stakes are also very prevalent, like Kistler v. Eightfold AI Inc., which treats AI scoring vendors as consumer reporting agencies.⁶³ Tailored, jurisdiction-specific compliance protocols are no longer optional; they are the only way to satisfy conflicting statutory mandates while minimizing enforcement liability.⁶⁴

IV. Conclusion

The convergence of labor rights and algorithmic accountability means organizations must adopt a unified strategy that covers compliance across all facets of workplace AI.⁶⁵ As courts and regulatory agencies increase scrutiny on automated systems, employers must take proactive measures to satisfy NLRA bargaining duties, draft clear disclaimers in handbooks to protect at-will status, require vendor oversight, enforce bias audits, strict confidentiality, and consent protocols for GAI tools.⁶⁶ When organizations establish a thorough HITL framework they ensure that AI tools increase productivity without exposing them to liability.⁶⁷ The future of the workplace is undeniably augmented, but the legal structure of the organization must remain human-centric.⁶⁸ Formalizing these protocols can successfully protect organizations from systemic liability while enabling a culture of responsible, defensible innovation.⁶⁹

Cover Photo by Igor Omilaev on Unsplash

Author

Jozy Allen

Jozy Allen is a 2L JD Candidate from Cincinnati, Ohio. She completed her Undergraduate and Graduate degrees at Central Michigan University, where she played Division I lacrosse. In addition to Law Review, Jozy is the President of the Business Law Society and most recently worked at Squire Patton Boggs as a Summer Associate. Her legal interests are primarily corporate transactional law, and in her free time she coaches lacrosse!

References

1
Jurassic Park, Amazon Prime, at 36:08 (Universal Pictures 1993). ↩︎
2
Joseph O’Keefe & Nicholas Franchetti, Artificial Intelligence in the Workplace: Best Practices, Lexis Prac. Guidance (database updated July 15, 2026), https://plus.lexis.com/document?pdmfid=1530671&pddocfullpath=%2Fshared%2Fdocument%2Fanalytical-materials%2Furn%3AcontentItem%3A68WN-XD81-JPP5-227N-00000-00&pdcontentcomponentid=500749&pdislparesultsdocument=true&prid=2849acee-84f2-4159-9d47-0336fcbf2ffd&crid=3de37527-d629-42f6-aed6-e0a6ac75db94&pdisdocsliderrequired=true&pdpeersearchid=c4aa9ad3-6598-414b-b248-d1dfcabb4d61-1&ecomp=bphtk&earg=sr0#/0a432e83-3d0b-4c15-8283-9cdbc9a3b480 [https://perma.cc/L8KF-MD9F]. ↩︎
3
Id.; Mobley v. Workday, Inc., 740 F. Supp. 3d 796, 804-08 (N.D. Cal. 2024). ↩︎
4
Id.; McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802-03 (1973) (McDonnell establishes the burden-shifting framework for individual employment discrimination claims under Title VII and the ADA and the National Labor Relations Board (NLRB) evaluates discrimination and retaliation claims under Section 8(a)(3) of the NLRA using the Wright Line standard.); see generally N.L.R.B. v. Wright Line, 662 F.2d 899 (1st Cir. 1981). ↩︎
5
Mobley, 740 F. Supp. at 806, 810-11. ↩︎
6
National Labor Relations Act § § 7, 8(d), 29 U.S.C. § § 157, 158(d) (2024); Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e-2 (2024); Americans with Disabilities Act of 1990, 42 U.S.C. § 12112 (2024); O’Keefe & Franchetti, supra note 2 (While the NLRA serves as the primary statutory check on labor relations and collective bargaining under Section 7, claims of individual employment discrimination stemming from AI decision-making fall under Title VII and the ADA, enforced federally by the Equal Employment Opportunity Commission (EEOC) or at the state level by agencies such as the Ohio Civil Rights Commission (OCRC)). ↩︎
7
National Labor Relations Act §§ 8(a)(5), 8(d), 29 U.S.C. §§ 158(a)(5), 158(d) (2024); First Nat’l Maint. Corp. v. NLRB, 452 U.S. 666, 674 (1981); Omaha Typographical Union No. 190 v. NLRB, 545 F.2d 1138, 1141 (8th Cir. 1976); O’Keefe & Franchetti, supra note 2. ↩︎
8
NLRA § 8(d), 29 U.S.C. § 158(d) (2024); First, 452 U.S. at 674-75; Omaha, 545 F.2d 1138, 1141 (8th Cir. 1976); O’Keefe & Franchetti, supra note 2. ↩︎
9
NLRA § 7, 29 U.S.C. § 157 (2024); Stericycle, Inc., 372 N.L.R.B. No. 113 at 1, 8 (2023); O’Keefe & Franchetti, supra note 2. ↩︎
10
First, 452 U.S. at 677, 681-82; Omaha, 545 F.2d at 1141; Fireboard Paper Prods. Corp. v. NLRB, 379 U.S. 203, 223 (1964) (Stewart, J., concurring); O’Keefe & Franchetti, supra note 2. ↩︎
11
NLRA §§ 8(a)(5), 8(d), 29 U.S.C. § 158(a)(5), 158(d) (2024); Omaha, 545 F.2d at 1141; O’Keefe & Franchetti, supra note 2. ↩︎
12
NLRA § 8 (d), 29 U.S.C. § 158(d) (2024); NLRB v. Am. Nat’l Ins. Co., 343 U.S. 395, 402-04 (1952). ↩︎
13
Id.; Omaha Typographical Union No. 190 v. NLRB, 545 F.2d 1138, 1142 (8th Cir. 1976). ↩︎
14
First Nat’l Maint. Corp. v. NLRB, 452 U.S. 666, 677-79 (1981) (e.g., a partial business closure or the initial decision to automate). ↩︎
15
Id. at 677-82; NLRA § 8(a)(5), 29 U.S.C. § 158(a)(5) (2024). ↩︎
16
Rochet, 136 N.L.R.B. 1294, 1297-98 (1962); NLRB v. Columbia Tribune Publ’g Co., 495 F.2d 1384, 1391 (8th Cir. 1974); Newspaper Printing Corp. v. NLRB, 625 F.2d 956, 964 (10th Cir. 1980); O’Keefe & Franchetti, supra note 2. ↩︎
17
Omaha, 545 F.2d at 1142. ↩︎
18
Id.; see also First, 452 U.S. at 685; NLRB v. Reed & Prince Mfg. Co., 205 F.2d 131, 134-35 (1st Cir. 1953). ↩︎
19
O’Keefe & Franchetti, supra note 2. ↩︎
20
O’Keefe & Franchetti, supra note 2 (e.g., automated productivity tracking, AI-driven shift allocation, and real-time performance monitoring). ↩︎
21
NLRA § 8(d), 29 U.S.C. § 158(d) (2024); O’Keefe & Franchetti, supra note 2. ↩︎
22
NLRA § 8(a)(5), 29 U.S.C. § 158(a)(5) (2024); NLRB v. Katz, 369 U.S. 736, 743-45 (1962); First Nat’l Maint. Corp. v. NLRB, 452 U.S. 666, 675 (1981); Omaha Typographical Union No. 190 v. NLRB, 545 F.2d 1138, 1141 (8th Cir. 1976); O’Keefe & Franchetti, supra note 2. ↩︎
23
NLRA § 8(a)(3), 29 U.S.C. § 158(a)(3) (2024); O’Keefe & Franchetti, supra note 2; see generally N.L.R.B. v. Wright Line, 662 F.2d 899 (1st Cir. 1981). ↩︎
24
Wright Line, 662 F.2d at 901-902; O’Keefe & Franchetti, supra note 2. ↩︎
25
O’Keefe & Franchetti, supra note 2. ↩︎
26
Id. ↩︎
27
NLRB Gen. Counsel Mem. 18-04, Guidance on Handbook Rules Post-Boeing (June 6, 2018). ↩︎
28
NLRB Gen. Counsel Mem. 23-02, Electronic Monitoring and Algorithmic Management of Employees Interfering with the Exercise of Section 7 Rights (Oct. 31, 2022). ↩︎
29
NLRB Gen. Counsel Mem. 25-05, Recission of Certain General Counsel Memoranda (Feb. 14, 2025); O’Keefe & Franchetti, supra note 2. ↩︎
30
O’Keefe & Franchetti, supra note 2. ↩︎
31
Daniel J. Alexander II, Representing Start-Ups and Emerging Companies § 8.10 ( 1st ed. 2026); O’Keefe & Franchetti, supra note 2. ↩︎
32
Id. ↩︎
33
Id.; Quijano v. Atkins-Kroll, Inc., 2008 Guam 14 (2008) (holding that handbook termination policies create implied contractual obligations); D’Angelo v. Gardner, 819 P.2d 206, 209-11 (Nev. 1991) (holding that employee reliance on acknowledged handbook terms modifies at-will employment). ↩︎
34
Alexander II, supra note 32. ↩︎
35
Watson v. Idaho Falls Consol. Hosps., 720 P.2d 632, 635-37 (1986); Alexander II, supra note 32. ↩︎
36
Alexander II, supra note 32; see generally Thompson v. City of Idaho Falls, 887 P.2s 1094 (Idaho Ct. App. 1994). ↩︎
37
S.C. Code Ann. § 41-1-110 (2024); see also Alexander II, supra note 32. ↩︎
38
Id. ↩︎
39
Alexander II, supra note 32 ↩︎
40
Stericycle Inc., 372 N.L.R.B.; O’Keefe & Franchetti, supra note 2. ↩︎
41
Id. ↩︎
42
O’Keefe & Franchetti, supra note 2 (e.g., productivity benchmarks, shift allocation systems, or electronic surveillance systems). ↩︎
43
NLRA § 8(a)(5), 29 U.S.C. § 158(a)(5); Stericycle Inc., 372 N.L.R.B; NLRB v. Katz, 369 U.S. 736, 743 (1962). ↩︎
44
NLRA § 8(d), 29 U.S.C. § 158(d); First Nat’l Maint. Corp. v. NLRB, 452 U.S. 666, 674 (1981). ↩︎
45
NLRA §§ 7, 8(a)(1), 29 U.S.C. §§ 157, 158(a)(5); Stericycle, 372 N.L.R.B. ↩︎
46
Id. ↩︎
47
See generally Stericycle, 372 N.L.R.B. ↩︎
48
Id. ↩︎
49
Mobley v. Workday, Inc., 740 F. Supp. 3d 796, 804-08 (N.D. Cal. 2024); Restatement (Third) of Agency § 1.01 (Am. L. Inst. 2006); O’Keefe & Franchetti, supra note 2. ↩︎
50
Mobley, 740 F. Supp. at 804-05. ↩︎
51
Id. at 804-05, 814. ↩︎
52
No. 1:23-cv-02102, 2024 U.S. Dist. LEXIS 84968 at 14 (D. Md. May 10, 2024), aff’d 2024 U.S. App. LEXIS 29655 (4th Cir. Nov. 21, 2024). ↩︎
53
Id.; Mobley, 740 F. Supp. at 809-11; O’Keefe & Franchetti, supra note 2. ↩︎
54
EEOC Decision No. 1:22-cv-02565 (E.D.N.Y. Aug. 9, 2023). ↩︎
55
Americans with Disabilities Act of 1990 §102(a), (b)(6), 42 U.S.C. § 12112(a), (b)(6) (2024); Mobley v. Workday, Inc., 740 F. Supp. 3d 796, 804 (N.D. Cal. 2024); 42 U.S.C. § 12112(a), (b) (2024). ↩︎
56
U.S. Equal Emp. Opportunity Comm’n, Algorithms, Artificial Intelligence, and Disability Discrimination in Hiring (May 12, 2022), https://www.ada.gov/resources/ai-guidance/ [] [https://perma.cc/2ZBC-XS7N]; O’Keefe & Franchetti, supra note 2; see generally, Mobley, 740 F. Supp. ↩︎
57
O’Keefe & Franchetti, supra note 2; Mobley, 740 F. Supp. 3d at 802-03. ↩︎
58
O’Keefe & Franchetti, supra note 2; U.S. Equal Emp. Opportunity Comm’n, supra note 67. ↩︎
59
820 Ill. Comp. Stat. 42/5 (2024); 775 Ill. Comp. Stat. 5/2-102(L)(2) (2026); N.Y.C. Local Law No. 144; Colo. Rev. Stat. §§ 6-1-1702, 6-1-1703 (2025); O’Keefe & Franchetti, supra note 2. ↩︎
60
820 Ill. Comp. Stat. 42/5 (2024); 775 Ill. Comp. Stat. 5/2-102(L)(2) (2026); Mobley, 740 F. Supp. at 812; O’Keefe & Franchetti, supra note 2. ↩︎
61
N.Y.C. Local Law No. 144, codified at N.Y.C. Admin Code §§ 20-870 (2021). ↩︎
62
Colo. Rev. Stat. §§ 6-1-1702, 6-1-1703 (2025); O’Keefe & Franchetti, supra note 2. ↩︎
63
Consumer Fin. Prot. Bureau, CFPB Circular on FCRA Application to AI Background Dossiers (2023); O’Keefe & Franchetti, supra note 2; Kistler v. Eightfold AI Inc., No. C26-00214 LEXIS (Cal. Super Ct. Contra Costa Cnty. Filed Jan. 20, 2026) (where AI screening vendors generating applicant scores were alleged to act as “consumer reporting agencies” and must be subject to mandatory disclosures and consent requirements). ↩︎
64
O’Keefe & Franchetti, supra note 2. ↩︎
65
Id. ↩︎
66
NLRA § 8(d), 29 U.S.C. § 158(d) (2024); Quijano v. Atkins-Kroll, Inc., 2008 Guam 14 (2008); S.C. Code Ann. § 41-1-110; Mobley v. Workday, Inc., 740 F. Supp. 3d 796, 804-08 (N.D. Cal. 2024); O’Keefe & Franchetti, supra note 2. ↩︎
67
O’Keefe & Franchetti, supra note 2. ↩︎
68
Id. ↩︎
69
Id. ↩︎