AI and the Age of Radical Transparency: When Reputation Becomes Verifiable – RESEARCH & PODCAST 2026

Published by Di Tran University (DTU), Research Division on AI Governance, Institutional Trust, and Proof-of-Work Systems, this paper is provided for educational and analytical purposes only. The analysis explores artificial intelligence as an emerging transparency and credibility infrastructure, grounded in publicly available research and cited sources. All referenced case studies are drawn from documented reports and academic literature; DTU does not assert independent investigative findings or legal conclusions beyond those sources. This publication is intended to advance scholarly dialogue on governance, verification, and ethical digital systems, and should be read as research commentary rather than legal, financial, or regulatory advice.

---

The global information ecosystem is currently undergoing a structural transformation, shifting from a model of reputation based on localized trust and institutional authority to one governed by algorithmic verification and radical transparency. This transition is catalyzed by the maturation of artificial intelligence as a large-scale pattern recognition and verification system.¹ In this emerging reality, AI functions as a “credibility amplifier,” a socio-technical force that increasingly exposes inconsistencies in human records, verifies professional expertise through multi-dimensional data analysis, and evaluates ethical behavior against stated principles.³ The core thesis of this shift is that as AI systems integrate more deeply into the digital infrastructure, they will reward authenticity and verifiable expertise while systematically exposing misinformation, professional exaggeration, or unethical conduct. This research report analyzes the mechanisms of this transformation, its psychological and social implications, the evolving governance landscape, and the strategic readiness required for the next decade of digital interaction.

1. Conceptual Framework

The conceptual foundation of AI-driven transparency is rooted in the transition from “outcome-based thinking” to “process-based thinking” regarding trust and reputation.³ Historically, digital reputation was perceived as a static output—a rating, a credential, or a curated biography. However, in the age of artificial intelligence, reputation is redefined as an emergent property of a layered system of data collection rules, verification standards, and visibility algorithms.³

1.1 AI as a Large-Scale Pattern Recognition and Verification System

Artificial intelligence systems, particularly those utilizing machine learning (ML) and deep learning (DL), possess the unique capability to identify relationship patterns across vast and distributed datasets that are invisible to human actors.² These systems do not merely “read” data; they perform complex operations such as prioritisation, classification, association, and filtering.⁶ By determining relationships between entities via semantic and connotative abilities, AI can detect whether an individual’s professional claims align with their public footprint, behavioral history, and peer-validated records.⁶

A critical conceptual distinction in this framework is the difference between “Internal Consistency” and “External Validity”.⁸ Sophisticated AI systems can create self-reinforcing loops of internal logic that appear impressive but fail when tested against independent reality. True transparency infrastructure requires that pattern recognition be regularly validated against external benchmarks and human judgment to prevent “defensive institutional responses” where algorithmic patterns are protected at the expense of technical accuracy.⁸

1.2 Defining the New Infrastructure of Trust

The shift toward verifiable reputation is characterized by several key conceptual pillars. Digital reputation refers to the confidence stakeholders place in information based on how it is collected, verified, and presented, rather than just the content itself.³ Algorithmic accountability is the process of holding the developers and operators of these systems responsible for ensuring that decisions result in fair and transparent outcomes.¹⁰ Furthermore, “Proof-of-Work Credibility” is an emerging model where trust is built through cryptographically signed, timestamped, and externally anchored records of actions, such as “refusal proofs” in AI safety or “content credentials” in digital media.¹¹

Concept	Definition in the AI Era	Institutional Impact
Digital Reputation	Confidence derived from systemic verification and data consistency.	Shifts value from “curated image” to “auditable history.”
Algorithmic Accountability	The duty of designers to provide evidence of system fairness and prevent harm.	Necessitates internal and external audit structures.
Proof-of-Work Credibility	Verifiable evidence of action (e.g., hash chains, cryptographic logs).	Reduces the power of unverified claims or “vaporware.”
Transparency Infrastructure	The underlying software (e.g., traceability systems) that centralizes intelligence.	Transforms stories from “told” narratives to “shown” evidence.

³

1.3 Historical Evolution of Reputation Systems

The transition from traditional to AI-driven verification represents a fundamental shift in how human societies manage risk and collaboration. Traditional reputation systems relied heavily on word-of-mouth, which was limited by subjectivity and labor-intensive manual checks.⁴ In contrast, AI-driven systems leverage predictive analytics and real-time anomaly detection to provide an agile and accurate assessment of credibility.⁴

While traditional methods were often compartmentalized and oriented toward historical data, AI systems are increasingly oriented toward “real-time transparency.” For example, where a university once verified a degree through a manual letter of confirmation, modern systems like Hyperstack utilize pattern recognition across metadata and blockchain-linked trust scores to validate credentials in seconds.¹⁵ This shift represents the move from authority-based trust—trusting a source because of its name—to evidence-based trust—trusting a source because its record is internally and externally consistent.

2. AI as a Credibility Engine

Artificial intelligence functions as a credibility engine by automating the appraisal and verification of human activity at a scale that was previously impossible. This is achieved through the integration of Large Language Models (LLMs), search indexing, and multi-stage reasoning pipelines that move beyond surface-level keyword matching.¹⁶

2.1 Consistency Evaluation Across Public and Professional Records

The core mechanism of AI as a credibility engine is its ability to evaluate consistency across public records, publications, and professional histories. Public records, ranging from birth certificates to court proceedings, are increasingly managed by AI systems that automate categorization and filing, ensuring that data is organized and searchable for both citizens and regulators.¹⁷ This automation reduces human error and enhances the speed of information retrieval, making it difficult for individuals to obscure past actions or contradictory professional claims.¹⁷

In the professional realm, AI systems utilize “Metadata Matching” and “Anomaly Detection” to validate expertise. For instance, AI models trained on known legitimate credentials can learn typical naming conventions, issuer formatting, and signature positioning.¹⁵ If a submitted document deviates significantly from these patterns—such as mismatched encoding or timestamp inconsistencies—it is flagged for human review.¹⁵ This capability is being extended to “Corporate Reputation Measurement,” where AI analyzes unstructured data from social media, news, and financial logs to identify shifts in stakeholder perception before they escalate into crises.⁵

2.2 The Role of LLMs in Detecting Contradictions

LLMs have transformed from simple text generators into sophisticated reasoning agents capable of identifying misinformation through multi-stage pipelines. Advanced research, such as that conducted at MIT, has demonstrated LLM-based systems that decompose content into explicit and implicit claims, identify rhetorical “framing devices,” and generate “missing context” questions.¹⁶

These systems retrieve evidence from external sources, such as the Google Fact Check API and semantic search engines, to synthesize grounded explanations.¹⁶ This process allows AI to detect “high-impact misinformation”—content that may be technically accurate but is decontextualized to mislead the reader.¹⁶ For example, in the job market, LLMs are used to compare a candidate’s resume against their LinkedIn footprint and external project records. Research shows that “factual consistency” in LLM outputs can be used as a benchmark for objectivity, exposing where a model—or a human actor—might be tailoring information to predetermined biases.¹⁸

2.3 Evidence of Exposure: Fraud, Expertise, and Conduct

Real-world applications of AI as a credibility amplifier have already begun to expose significant misconduct. In the scientific community, AI audits of journals have identified large-scale plagiarism and data falsification, leading to the retraction of hundreds of papers in “mega-journals” like Heliyon.²⁰ In corporate environments, AI-driven anomaly detection has uncovered internal accounting frauds, such as the 2025 Macy’s scandal where an employee concealed $154 million in delivery expenses over three years to inflate profits and bonuses.²¹

The hiring process has become a frontline for AI-driven verification. Companies like Malwarebytes and startups like Vidoc Security have documented the rise of “fake applicants” who use AI-generated resumes and deepfake video filters to pass initial screenings.²² AI-based detection methods now look for telltale visual cues of real-time filters—such as unnatural blinking or distorted edges when a hand is placed in front of the face—as well as behavioral red flags like “long, noticeable pauses” that suggest off-screen coaching.²²

Case Study	Misconduct Type	AI/Data Mechanism of Exposure
Macy’s (2025)	Accounting Fraud ($154M)	Internal audit of delivery expense misclassification.
Vidoc Security	Fake Engineering Candidates	Visual “hand-to-face” test exposing AI deepfake filters.
Heliyon Audit	Academic Plagiarism	Automated similarity and internal audit of paper metadata.
Malwarebytes	Resume Fraud	Detecting “flawless” AI resumes with zero project substance.

²⁰

3. Ethical Amplification vs. Exposure

The dual effect of AI as a credibility amplifier is its capacity for “Ethical Amplification”—making virtuous actions more impactful—and “Exposure”—making deceptive actions more costly.²⁴ This dynamic creates a “transparency dividend” for ethical organizations while increasing the “deterrence deficit” for dishonest actors.²⁵

3.1 Benefits to Ethical Individuals and Organizations

Ethical organizations leverage AI through “Traceability Software” and “Transparency Infrastructure.” These systems serve as the “quiet backbone” that turns scattered updates into a living record of responsible behavior.¹³ In the sustainability sector, ethical amplification ensures that initiatives are grounded in justice and equity, using data to show—rather than just tell—a brand’s story.²⁴

Traceability software allows brands to:

1. Provide tangible proof of origins at the point of sale.
2. Shorten the response time for buyers and regulators through verified records.
3. Align internal teams (sourcing, marketing, legal) around a single “source of truth.”
4. Track supplier performance and environmental impact in real-time, moving from reactive reporting to proactive shaping of outcomes.¹³

3.2 Exposing Dishonesty and Inconsistent Narratives

For those engaged in “AI washing” or professional exaggeration, the era of radical transparency is inherently threatening. AI washing—the practice of misrepresenting a firm’s AI capabilities for financial gain—is increasingly detected by systems that analyze the gap between branding and technical reality.¹ These “deceptive or overstated practices” might offer short-term legitimacy but result in severe long-term reputational damage and erosion of trust.¹

Furthermore, AI creates a record that is difficult to “un-see.” In social media and public discourse, algorithms can unintentionally worsen disagreements by prioritizing engagement, but they also create a searchable trail of statements that can be cross-referenced for inconsistency.²⁶ The “liar’s dividend”—the ability of a person to claim that real footage of their misconduct is a deepfake—is being curtailed by “Content Credentials” and cryptographic proof systems that verify the origin of authentic media.¹²

3.3 The Risks of False Positives and Algorithmic Bias

Despite its potential, the use of AI as a moral judge introduces significant risks. Algorithmic bias occurs when computer systems produce “systematic and repeatable decisions” that create unfair or discriminatory outcomes.¹⁰

1. The Hidden Bias Problem: AI models trained on historical data may learn to favor “top-tier” schools or specific demographics, flagging credentials from lesser-known institutions as “low trust” simply because they appear less frequently in the training set.¹⁵
2. Context Blindness: Algorithms lack a moral “conscience.” They can identify a technical violation—such as shared SIM cards in a mobile salary system—but fail to recognize it as a “humanitarian necessity” in remote regions.⁹
3. Model Collapse: Continuous training of AI models on AI-generated data can lead to a degradation in quality and accuracy over time, potentially corrupting the historical record and creating “digital heaps” of unmanaged, unreliable information.²⁷

4. The End of Information Asymmetry

Information asymmetry, once a defining feature of professional and institutional power, is being eroded as AI democratizes the ability to verify claims and analyze complex data. This transition shifts the locus of power from those who possess information to those who can verify it.

4.2 Reducing the Gap Between Insiders and the Public

AI reduces the gap between insiders and the general public by providing tools for “distance viewing” and “close reading” of massive datasets.²⁸ For example, in the financial sector, AI improves “Financial Reporting Quality” (FRQ) by reducing manual error and subjectivity, thereby aligning the interests of management (agents) and investors (principals).²⁹ In public administration, AI enables citizens to query vast archives of government records in plain language, fostering an informed citizenry and reinforcing democratic principles of accountability.¹⁷

4.2 Implications for Professionals and Educators

For professionals—including doctors, lawyers, and engineers—the “signaling value” of traditional credentials is changing. Research from Princeton and Dartmouth indicates that the availability of LLMs has reduced the value of written communication as a signal of high ability, as both top-tier and lower-tier candidates can now produce polished materials.³⁰ This necessitates a shift toward “documented outcomes,” where professionals must prove their expertise through auditable data rather than just credentials.

In education, AI feedback systems are proving as effective as human feedback in many contexts, offering the potential for timely, data-driven interventions.³¹ However, this also introduces the risk of “productivity theater,” where students or employees use AI to simulate effort, forcing a transition toward evaluation models that prioritize the “process of creation” over the “final output.”

4.3 Transitioning from Authority-Based to Evidence-Based Trust

We are moving toward a trust model based on “verifiable behavior” rather than “claims of utility”.¹² In this model, trust is not a static number or a prestigious title; it is a constructed reality supported by identity-linked accounts, auditable moderation processes, and traceable histories.³

Trust Dimension	Authority-Based Model	Evidence-Based (AI) Model
Source of Truth	Institutional affiliation (e.g., Ivy League).	Cross-platform data consistency.
Verification Basis	Human-to-human reference checks.	Algorithmic metadata and pattern matching.
Longevity	Episodic (renewed via titles/awards).	Continuous (maintained via real-time audit).
Transparency	Low (Internal decisions are “black box”).	High (Decisions are logged and auditable).

³

5. Psychological and Social Implications

The psychological perception of AI as a global “judge” creates significant tension as societies adapt to a reality where one’s “digital shadow” is perpetually examined for flaws.

5.1 The Perception of Algorithmic “Judgment”

Many individuals perceive this shift as a form of relentless machine-based judgment. Unlike human judgment, which may offer empathy or contextual grace, algorithmic systems are often seen as “accurate but heartless”.⁹ When decisions appear to come from “the system” rather than people, the language of accountability shifts from “I approved it” to “the system processed it,” leading to a diffusion of responsibility that many find unsettling.⁹

5.2 Fear, Resistance, and Technological Transitions

History provides clear parallels for the current resistance to AI. When steam locomotives began operating in the 1820s, physicians claimed the velocity would cause “brain damage” or “insanity,” reflecting a fear of unnatural speed that mirrors current anxieties about the speed of algorithmic decision-making.³³ Similarly, photography in the 1830s was met with “moral panic” over its ability to capture reality without artistic interpretation, much like modern fears about deepfakes and the “death of authenticity”.³³

This resistance is often an “institutional defensive response.” When sophisticated pattern recognition becomes entrenched in an organization, any criticism of those patterns is treated as an attack on the organization’s competence.⁸ This leads to a “deterrence deficit,” where individuals might fear the algorithm more than they respect the underlying ethical principle.²⁵

5.3 Adaptation and “Productivity Theater”

As AI becomes the tool for measuring performance, humans are adapting through “productivity theater”—the simulation of work to satisfy an algorithm. Examples include the use of “mouse jigglers” to bypass employee monitoring or the Bombardment of hiring systems with AI-optimized resumes.³⁰ This behavior suggests that while AI can amplify credibility, it can also incentivize sophisticated forms of deception designed to meet the metric without providing the value.

6. Governance, Ethics, and Risks

As the power to define reputation shifts to machines, the governance of those machines becomes the primary ethical challenge of the decade. This requires a transition from voluntary “best practices” to binding legal requirements.

6.1 Privacy Concerns and the Surveillance Dilemma

The move toward radical transparency inherently threatens individual privacy. AI systems that process 1.6 million patient records for health triage (as in the DeepMind/NHS case) or scrape billions of images for facial recognition (Clearview AI) raise fundamental questions about consent and “data sovereignty”.³⁴ The “Privacy Puzzle” is further complicated by the need for “Refusal Proofs”—logs that prove an AI system blocked harmful content. While these logs are essential for auditing, they create a new category of sensitive data that could be altered or misused.¹¹

6.2 Algorithmic Bias and Reputational Harm

The “Black Box” problem remains a significant hurdle. If an AI system makes a decision that leads to the unauthorized exposure of personal data or a discriminatory hiring outcome, it is difficult to determine whether the fault lies with the developer, the data, or the user.³⁴ This is particularly problematic in “High-Risk” sectors such as:

• Hiring and Screening: Rejection of candidates based on biased historical data.³⁶
• Creditworthiness: Lowering limits for specific demographics despite similar financial profiles.⁹
• Law Enforcement: False positives leading to wrongful arrests.³⁸

6.3 Responsible AI Governance Frameworks

Global regulators are responding with frameworks designed to ensure transparency and accountability.

• EU AI Act: Categorizes AI systems into risk tiers (Prohibited, High, Limited, Minimal). It mandates that “High-Risk” systems—such as those used in critical infrastructure or recruitment—conduct rigorous data governance and maintain unalterable logs.³⁹
• NIST AI RMF: A voluntary framework that helps organizations “Govern, Map, Measure, and Manage” AI risk, focusing on trustworthiness traits like fairness and explainability.⁴¹
• Algorithmic Accountability Act of 2022 (US): Requires companies to assess the risk of their algorithms for bias and transparency, submitting reports to the FTC for oversight.¹⁰

Framework	Tier/Function	Key Requirement	Penalty for Non-Compliance
EU AI Act	Unacceptable Risk	Outright prohibition of social scoring and untargeted scraping.	Up to €35M or 7% of global revenue.
EU AI Act	High Risk	Data governance, human oversight, and technical documentation.	Up to €15M or 3% of global revenue.
NIST RMF	Map & Measure	Identifying context-specific risks and validating performance.	Voluntary (No direct fines, but market risk).
ISO 42001	Management System	Documented evidence of AI governance controls.	Loss of certification/trust.

³⁹

7. Future Outlook (2026–2035)

The next decade will see the normalization of “continuous reputation auditing.” By 2030, reputation will no longer be a periodic assessment but a “live stream” of verified actions.³²

7.1 AI as Continuous Reputation Auditing

By 2030, AI technologies will transform external auditing by enabling “full-population risk analysis.” Instead of testing samples, AI will audit 100% of a company’s transactions, boosting fraud detection accuracy to over 85%.³² This “networked audit system” will cut manual reconciliation by up to 90%, making “reputational anomalies” immediately visible to investors and regulators.²⁹

7.2 The Emergence of “Proof-of-Work” Credibility Models

Reputation will increasingly depend on “Know Your Agent” (KYA) standards and “Cryptographic Proofs.” As AI agents begin to control funds and execute transactions without human intervention, verifiable credentials will be required to prove an agent’s authority and adherence to ethical rules.⁴⁴ This “transparency infrastructure” will move from a differentiator to a baseline expectation, much like the “Proof of Reserves” shift in the crypto industry following the FTX collapse.⁴⁴

7.3 How Institutions and Professionals Should Prepare

The future belongs to those who master “dual literacy”—the ability to read both the code (algorithmic logic) and the room (human relationships).⁴⁵

Predictions for 2030:

1. Spending Growth: Organizations will quadrupedal their spending on AI governance, reaching over $1 billion by 2030 as fragmented global regulations take hold.⁴³
2. Shift in Skills: Auditors and compliance officers will need to transition from “inspecting numbers” to “interpreting algorithmic logic”.⁹
3. The Rise of Digital Product Passports: Consumers will demand a “product-level story” that is coherent across all channels, from raw material to finished goods.¹³

8. Practical Readiness Guide

For individuals and organizations to survive the transition to radical transparency, they must treat their “digital record” as their most valuable asset.

8.1 Consistency of the Public Record

AI systems excel at detecting “fragmented data” and “inconsistencies”.⁴

• Action: Conduct a “semantic audit” of all public platforms. Ensure that historical records, LinkedIn profiles, and corporate bios do not contain timestamp or achievement discrepancies.
• Action: Synchronize internal dashboards with external narratives. If your sustainability report claims a 20% reduction in waste, ensures that supply chain logs (e.g., in traceability software) provide the “scannable evidence” to support it.¹³

8.2 Documented Outcomes Over Claims

As LLMs make “polished communication” a commodity, actual data becomes the only way to differentiate yourself.³⁰

• Action: Shift from “resume-based” hiring to “performance-based” verification. Require “live, interactive participation” and “identity verification” at the start of any high-stakes professional interaction.²²
• Action: Adopt “Proof-of-Work” tools. Use blockchain-linked credentials or digital signatures to anchor your professional achievements in an immutable record.¹¹

8.3 Ethical Operations and Governance

Reputation is no longer just about “avoiding scandal”; it is about “demonstrating governance”.⁹

• Action: Implement an “AI Integration Capability Model” (AICM) to ensure interoperable data and ethical infrastructure.⁷
• Action: Establish clear “Separation of Duties” (SoD). Ensure that the people who approve actions are not the same people who record them in the system, preventing the “Macy’s-style” manipulation of records.²¹

8.4 Verifiable Expertise

In the era of deepfakes, “authenticity” must be provable through physical and digital markers.

• Action: Use “Content Credentials” (C2PA) for all high-value media creation, ensuring that your content can be verified back to its origin.¹²
• Action: Develop a “human checkpoint” system. Use AI for large-scale screening, but ensure that “low-trust” flags are examined thoughtfully by humans who can account for “pedagogical and contextual nuances”.¹⁵

Conclusion

Artificial intelligence is not merely a tool for efficiency; it is becoming the world’s most powerful “credibility amplifier.” By making human patterns transparent, it rewards those whose actions align with their words while systematically eroding the influence of those who rely on misinformation or exaggeration. However, this shift toward radical transparency is a socio-technical challenge that requires more than just better algorithms; it requires a new ethics of governance, a commitment to data integrity, and a psychological adaptation to the end of information asymmetry. As we move toward 2035, the most successful individuals and institutions will be those who recognize that in an AI-driven world, reputation is no longer what you claim to be—it is what you can prove you have done. ¹

Works cited

1. AI Washing and the Erosion of Digital Legitimacy: A Socio-Technical Perspective on Responsible Artificial Intelligence in Business – arXiv, accessed February 19, 2026, https://arxiv.org/html/2601.06611v1
2. Reputation and Accountability in the Age of Algorithms | Institute for Public Relations, accessed February 19, 2026, https://instituteforpr.org/reputation-and-accountability-in-the-age-of-algorithms/
3. How Digital Trust Is Constructed: Online Reputation Explained – Equinox Cleaning, accessed February 19, 2026, https://equinoxcleaning.net/how-digital-trust-is-constructed-framework/
4. (PDF) A comparative analysis between AI and traditional methods in management controlA comparative analysis between AI and traditional methods in management control – ResearchGate, accessed February 19, 2026, https://www.researchgate.net/publication/392517424_A_comparative_analysis_between_AI_and_traditional_methods_in_management_controlA_comparative_analysis_between_AI_and_traditional_methods_in_management_control
5. Algorithms and accountability: Navigating corporate reputation in the digital age, accessed February 19, 2026, https://www.agilitypr.com/pr-news/branding-reputation/algorithms-and-accountability-navigating-corporate-reputation-in-the-digital-age/
6. ALGORITHMIC ACCOUNTABILITY – World Wide Web Foundation, accessed February 19, 2026, https://webfoundation.org/docs/2017/07/Algorithms_Report_WF.pdf
7. Integrating AI in Public Governance: A Systematic Review – MDPI, accessed February 19, 2026, https://www.mdpi.com/2673-6470/5/4/59
8. Pattern Recognition vs Validation: When Smart Systems Get Trapped – VerityAI, accessed February 19, 2026, https://verityai.co/blog/pattern-recognition-validation-science-ai-systems
9. AI Audits Numbers, Not Ethics: Why Humans Must Govern …, accessed February 19, 2026, https://www.corporatecomplianceinsights.com/ai-audits-numbers-not-ethics-humans-must-govern/
10. Ask the experts – What is algorithmic accountability? – Boston University, accessed February 19, 2026, https://www.bu.edu/hic/2022/02/15/ask-the-experts-what-is-algorithmic-accountability/
11. The Accountability Gap: Why AI Systems Need Cryptographic Proof of What They Refused to Generate | by VeritasChain Standards Organization (VSO) – Medium, accessed February 19, 2026, https://medium.com/@veritaschain/the-accountability-gap-why-ai-systems-need-cryptographic-proof-of-what-they-refused-to-generate-8a64d9978cde
12. The State of Content Authenticity in 2026, accessed February 19, 2026, https://contentauthenticity.org/blog/the-state-of-content-authenticity-in-2026
13. Why Traceability Software Is the Infrastructure Your Brand Needs, accessed February 19, 2026, https://www.retraced.com/blogs/magazine/why-traceability-software-is-the-infrastructure-your-brand-needs
14. AI-driven corporate reputation measurement in digital ecosystems: A …, accessed February 19, 2026, https://pubmed.ncbi.nlm.nih.gov/41197361/
15. The AI Behind Smarter Credentials: How Verification, Trust, and …, accessed February 19, 2026, https://thehyperstack.com/blog/ai-digital-credentials-verification/
16. Multi-Stage LLM Reasoning for Automated … – DSpace@MIT, accessed February 19, 2026, https://dspace.mit.edu/bitstream/handle/1721.1/164663/nair-naira-meng-eecs-2025-thesis.pdf?sequence=1&isAllowed=y
17. The Impact of AI on Public Records Transparency | RecordsKeeper.AI, accessed February 19, 2026, https://www.recordskeeper.ai/ai-public-records-transparency/
18. ConsistencyAI: A Benchmark to Assess LLMs’ Factual Consistency When Responding to Different Demographic Groups – arXiv, accessed February 19, 2026, https://arxiv.org/html/2510.13852v1
19. Linguistic features of AI mis/disinformation and the detection limits of LLMs – PMC, accessed February 19, 2026, https://pmc.ncbi.nlm.nih.gov/articles/PMC12800167/
20. Retraction Watch – Tracking retractions as a window into the scientific process, accessed February 19, 2026, https://retractionwatch.com/
21. Macy’s $154M Lesson: Why Every Company Needs Separation of Duties – LogicManager, accessed February 19, 2026, https://www.logicmanager.com/resources/corporate-governance/macys-154m-lesson-why-every-company-needs-separation-of-duties/
22. Deepfakes, AI resumes, and the growing threat of fake applicants …, accessed February 19, 2026, https://www.malwarebytes.com/blog/inside-malwarebytes/2025/12/deepfakes-ai-resumes-and-the-growing-threat-of-fake-applicants
23. AI and the Rise of Resume Fraud – NPAworldwide, accessed February 19, 2026, https://npaworldwide.com/blog/2025/05/16/ai-and-the-rise-of-resume-fraud/
24. Ethical Amplification → Area → Resource 1 – Lifestyle → Sustainability Directory, accessed February 19, 2026, https://lifestyle.sustainability-directory.com/area/ethical-amplification/resource/1/
25. Deepfakes and the Rise of Bad Behavior: How AI Enables Evasion of Accountability, Erodes Trust, and Challenges Moral Psychology in Crimes from Workplace Misconduct to Million-Dollar Scams – Reddit, accessed February 19, 2026, https://www.reddit.com/r/psychology/comments/1n81r18/deepfakes_and_the_rise_of_bad_behavior_how_ai/
26. Media Amplification of Condemnation → Area → Resource 1 – Lifestyle → Sustainability Directory, accessed February 19, 2026, https://lifestyle.sustainability-directory.com/area/media-amplification-of-condemnation/resource/1/
27. LLMs as Historical Actors: How AI Systems Influence the Web’s Evolution – Macquarie University, accessed February 19, 2026, https://research-management.mq.edu.au/ws/portalfiles/portal/442423721/441403304.pdf
28. AI to review government records: new work to unlock historically significant digital records, accessed February 19, 2026, https://pmc.ncbi.nlm.nih.gov/articles/PMC12442487/
29. The Role of Artificial Intelligence in Enhancing Financial Reporting Quality: Evidence from Saudi Arabia’s Vision 2030 Transformation – David Publishing, accessed February 19, 2026, https://www.davidpublisher.com/Public/uploads/Contribute/68d5e8e9d2b71.pdf
30. How AI amplifies resume fraud and other job seeker cheating – TechTarget, accessed February 19, 2026, https://www.techtarget.com/searcherp/podcast/How-AI-amplifies-resume-fraud-and-other-job-seeker-cheating
31. Full article: How does artificial intelligence compare to human feedback? A meta-analysis of performance, feedback perception, and learning dispositions – Taylor & Francis, accessed February 19, 2026, https://www.tandfonline.com/doi/full/10.1080/01443410.2025.2553639
32. The Future of Auditing: How AI Will Transform the Profession by 2030 – ResearchGate, accessed February 19, 2026, https://www.researchgate.net/publication/396736684_The_Future_of_Auditing_How_AI_Will_Transform_the_Profession_by_2030
33. A Historical Perspective on the Pushback Against AI, accessed February 19, 2026, https://etcjournal.com/2025/10/24/a-historical-perspective-on-the-pushback-against-ai/
34. Data Breaches and Liability in the Age of AI: Who’s responsible? – The Barrister Group, accessed February 19, 2026, https://thebarristergroup.co.uk/blog/ai-data-breaches-and-liability-whos-responsible
35. The Ethics of AI in Cybersecurity – Insights2TechInfo, accessed February 19, 2026, https://insights2techinfo.com/the-ethics-of-ai-in-cybersecurity/
36. From Algorithms to Deepfakes: AI Risks Every Employer Must Confront | News & Events, accessed February 19, 2026, https://www.clarkhill.com/news-events/news/from-algorithms-to-deepfakes-ai-risks-every-employer-must-confront/
37. Real-World Compliance Failures (Case Studies) | by Smrutisomyak | Jan, 2026 | Medium, accessed February 19, 2026, https://medium.com/@smrutisomyak/real-world-compliance-failures-case-studies-c45c0807dc6b
38. Risk Management Profile for AI and Human Rights – United States Department of State, accessed February 19, 2026, https://2021-2025.state.gov/risk-management-profile-for-ai-and-human-rights/
39. How the EU AI Act Impacts US Businesses – CompliancePoint, accessed February 19, 2026, https://www.compliancepoint.com/privacy/how-the-eu-ai-act-impacts-us-businesses/
40. AI Compliance: Risk Management for Artificial Intelligence – Osano, accessed February 19, 2026, https://www.osano.com/articles/what-is-ai-compliance
41. NIST vs EU AI Act: Which AI Risk Framework Should You Follow? – MagicMirror, accessed February 19, 2026, https://www.magicmirror.team/blog/nist-vs-eu-ai-act-which-ai-risk-framework-should-you-follow
42. NIST AI Risk Management Framework: A simple guide to smarter AI governance – Diligent, accessed February 19, 2026, https://www.diligent.com/resources/blog/nist-ai-risk-management-framework
43. Global AI Regulations Fuel Billion-Dollar Market for AI Governance Platforms – Gartner, accessed February 19, 2026, https://www.gartner.com/en/newsroom/press-releases/2026-02-17-gartner-global-ai-regulations-fuel-billion-dollar-market-for-ai-governance-platforms
44. 7 Crypto Audit Industry Predictions for 2026 – LedgerLens, accessed February 19, 2026, https://ledgerlens.io/7-crypto-audit-industry-predictions-for-2026
45. Age of the Algorithm CIPR, accessed February 19, 2026, https://www.cipr.co.uk/CIPR/Network/Groups_/Public_Affairs/Blogs/Age_of_the_Algorithm.aspx
46. Data Quality Matters Most, but Can We Detect Contradictions During Ingestion? – Reddit, accessed February 19, 2026, https://www.reddit.com/r/Rag/comments/1q8e91i/data_quality_matters_most_but_can_we_detect/