Bias is one of those words that arrives with its own weather. People hear it and brace for a storm: ideological bias, cultural bias, political bias. But in the algorithmic age, “bias” is something both simpler and stranger. It is the tilt that emerges whenever we ask a machine to compress the world into patterns we can use.

Bias isn’t a flaw. It’s a property. Everything that filters reality—eyes, memory, law, language—bends it a little. But AI makes that bending rigid.

For arbitrators navigating disputes that increasingly involve AI-generated evidence, AI-assisted analysis, or questions about algorithmic decision-making, understanding how models bend reality is no longer optional. It is foundational literacy. And like any literacy, it begins with learning to see what is already there.

What Bias Actually Means in AI

In AI, bias is not an opinion. It’s a structural feature. A statistical lean. A tendency wired into a model by what it has seen, what it has ignored, and what its designers have shaped it to avoid.

Three categories matter most to arbitrators:

Data Bias: The Inherited Absence

AI learns from human traces: text, legal documents, case law, news archives, academic papers. If parts of reality are missing from that archive, the model inherits the absence. If one jurisdiction, language, or legal tradition is over-represented, the model inherits the surplus.

Example in practice: An arbitrator using AI to summarize case law asks for precedents on force majeure in construction contracts. The model returns 47 cases—46 from common law jurisdictions, one from civil law. Not because civil law cases don’t exist, but because English-language legal databases dominate the training data. The bias isn’t malicious; it’s statistical. But to an arbitrator adjudicating an international construction dispute under Swiss law, the imbalance could be dispositive.

Research by Bommasani et al. (2021) in “On the Opportunities and Risks of Foundation Models” documents how training data composition directly shapes model outputs, with English-language content representing approximately 92% of training data in major models despite English speakers comprising only 17% of the global population.

Design Bias: The Philosophical Inheritance

Models don’t just absorb data. They inherit the philosophies of their makers. Every model has an internal logic—helpful, cautious, experimental, fast, bold, conservative. Safety tuning, reinforcement learning, guardrails: these are design decisions that shape the personality of the system.

Example in practice: An arbitrator asks two different models to assess whether certain communications constitute an anticipatory breach. Model A hedges extensively, noting seventeen reasons the question cannot be definitively answered without additional context. Model B provides a confident three-paragraph analysis. Neither is “wrong”—they’ve been tuned for different risk profiles. Model A was trained to avoid overconfidence; Model B to minimize user friction. The arbitrator who doesn’t recognize this difference might mistake caution for thoroughness or confidence for competence.

Interaction Bias: The Dance Between User and System

AI is unusually impressionable. Your phrasing, tone, order of questions, or even your uncertainty can nudge a model into a particular shape. Each prompt is a small gravitational pull.

Example in practice: An arbitrator asks: “Can you help me understand the weaknesses in the claimant’s position?” versus “Can you analyze the strengths and weaknesses of both parties’ positions?” The first prompt creates an adversarial frame; the model will search harder for flaws. The second invites balance. Same facts, same model, radically different analytical orientation.

Perez et al. (2022) in “Discovering Language Model Behaviors with Model-Written Evaluations” demonstrate that prompt framing can shift model outputs by up to 40% on factual accuracy tasks, with leading questions producing systematically different responses than neutral queries.

How Bias Manifests in Arbitration Contexts

Bias doesn’t announce itself. It appears in the seams. For arbitrators, five patterns matter most:

Jurisdictional Defaults

Models trained primarily on U.S. and EU legal materials may reflexively apply common law reasoning even when analyzing civil law questions. Ask about “discovery” and the model assumes American-style depositions unless explicitly redirected.

Temporal Compression

Models flatten legal evolution. Ask about the enforceability of arbitration agreements and the model may synthesize cases from 1995 and 2023 without clearly marking the doctrinal shift that occurred in between.

Consensus Bias

Models gravitate toward majority positions. On contested questions — say, the proper interpretation of Article 28 of the UNCITRAL Model Law — the model will favor the interpretation that appears most frequently in its training data, which may or may not reflect the best reasoned position or the position adopted in the relevant jurisdiction.

Hallucination Patterns

Different models fill knowledge gaps differently. A bold model will confidently invent case citations. A cautious model will refuse to answer. Both are forms of bias, one toward saying too much, the other toward saying too little. For an arbitrator relying on AI research assistance, the difference is critical.

Maynez et al. (2020) in “On Faithfulness and Factuality in Abstractive Summarization” found that 70% of neural abstractive summaries contained factual inconsistencies, with legal and technical domains showing higher error rates than general text.

Cultural Translation Failures

Legal concepts don’t translate cleanly. “Good faith” in American contract law is not bonne foi in French law is not Treu und Glauben in German law. Models trained to find equivalences will compress these distinctions unless actively prevented from doing so.

A Brief Aside: Cutting on the Bias

In dressmaking, cutting on the bias means slicing the fabric diagonally, at forty-five degrees to the weave. Do this, and something curious happens. The cloth gains stretch. It drapes differently. It moves with the body instead of resisting it. The same material, the same fibers, the same threads but the angle changes everything.

AI bias works a bit like that.

A model is made of data the way a fabric is made of threads. Straight-grain answers are the ones that follow the warp and weft of the training set. They’re sturdy, predictable, and occasionally stiff. But tilt the question, twist the context, or introduce a structural assumption and suddenly the model stretches. It drapes. It behaves differently.

Every model has its own “grain,” its own hidden geometry. Some fall with a clean line. Some pull unexpectedly at the seams. Some cling; some glide.

And just as an experienced seamstress learns to feel the grain through her fingertips, AI-literate arbitrators learn to sense where a model has stretch and where it has none. You learn to predict how it will fall. You learn when a straight cut will hold its shape and when cutting on the bias will give you something fluid.

Bias, in both senses, is not about right or wrong. It's about understanding the fabric you're working with. Once you understand it, you know what to expect.

The Institutional Architectures of Bias

Different organizations optimize for different values. These institutional priorities echo through every output.

OpenAI (GPT-4, GPT-o1)

Philosophy: Clarity, broad accessibility, moderate risk tolerance
Strength: Structured analytical reasoning, clear explanations
Bias pattern: Gravitates toward consensus positions, hedges on politically contested questions
Arbitration relevance: Reliable for procedural analysis, less useful for novel jurisdictional questions

Anthropic (Claude)

Philosophy: Constitutional AI, ethical awareness, transparency
Strength: Nuanced reasoning on ambiguous questions, explicit about limitations
Bias pattern: Over-cautious on questions involving harm, may refuse helpful analysis
Arbitration relevance: Excellent for multi-jurisdictional analysis, may need coaxing on adversarial content

Meta (Llama)

Philosophy: Open source, developer control, minimal restrictions
Strength: Adaptable, can be fine-tuned for specialized domains
Bias pattern: Highly variable depending on fine-tuning; less corporate safety filtering
Arbitration relevance: Best for specialized deployments; requires careful evaluation

Google (Gemini)

Philosophy: Multi-modal integration, enterprise focus
Strength: Handles complex document analysis across formats
Bias pattern: Optimized for business contexts, less granular legal reasoning
Arbitration relevance: Useful for document review, less reliable for doctrinal analysis

[Note These are general tendencies and fine-tuning, version updates, and deployment contexts can affect behavior.]

When you choose a model, you’re not choosing a tool. You’re choosing a worldview, a set of institutional priorities, and a philosophy of risk. An arbitrator who doesn’t understand this distinction is flying blind.

Practical Techniques for Working with Bias

Bias is not a dragon to be slain. It is a landscape to learn to navigate. Here are specific techniques arbitrators can use:

Technique 1: Prompt for Provenance

Don’t ask: “What does the case law say about waiver of arbitration rights?”

Ask: “What does the case law in [specific jurisdiction] say about waiver of arbitration rights? Please distinguish between pre-2015 and post-2015 cases, and identify any circuit splits or doctrinal shifts.”

This prompt forces the model to make its assumptions visible.

Technique 2: Adversarial Prompting

After receiving an initial analysis, ask:

• “What would be the strongest counterargument to this position?”

• “What assumptions are you making that might not hold in [civil law/common law/hybrid] systems?”

• “What edge cases or exceptions did you not mention?”

• “If you were representing the opposing party, how would you attack this reasoning?”

Welleck et al. (2023) in “Generating Sequences by Learning to Self-Correct” show that adversarial follow-up prompts reduce factual errors by 35% and increase reasoning depth in legal analysis tasks.

Technique 3: Multi-Model Triangulation

Use different models as checks on each other. Where they agree, you likely have consensus positions. Where they diverge, you’ve found the contested ground—which is often exactly what matters most in arbitration.

Technique 4: Explicit Framing

Begin prompts with: “You are an expert in [specific legal domain]. Your training data may over-represent [jurisdiction/legal tradition]. Be especially careful about...”

This metacognitive framing activates different parts of the model’s knowledge graph.

Technique 5: Few-Shot Examples

Provide 2-3 examples of the kind of analysis you want before asking your actual question. Models are remarkably good at pattern matching. Show them the pattern you need.

Example:

Here's an example of the level of detail I need:

Question: Is an arbitration clause in a franchise agreement enforceable if the franchisee is an individual?
Good answer: Under Swiss law, yes, but the clause must meet additional transparency requirements under Article 179(1) CO...

Now please answer this question: [your actual question]

Technique 6: Document Model Provenance

In any written decision or legal memorandum that relies on AI assistance, maintain a record of:

• Which model and version you used

• The specific prompts you provided

• When the analysis was generated

• What verification steps you took

This isn’t just good practice—as AI-assisted analysis becomes more common in arbitration, it will likely become a disclosure obligation.

Why This Matters for Arbitrators

Three scenarios make AI bias particularly salient in arbitration:

Scenario 1: Evaluating AI-Generated Evidence

A party submits contract language they claim was “drafted by AI” or damage calculations “verified by AI analysis.” An arbitrator must assess not just whether the output is accurate, but whether the AI’s biases might have shaped the result in legally significant ways.

Scenario 2: Using AI as Research Assistant

You’re deciding a complex jurisdictional question involving the interaction of three different national laws. You use AI to survey the relevant precedents. The model’s jurisdictional bias becomes your jurisdictional bias—unless you actively correct for it.

Scenario 3: Assessing Algorithmic Decision-Making

The dispute itself involves an algorithm: credit scoring, employment screening, pricing systems. To evaluate whether the algorithm was discriminatory or violated contractual obligations, you must understand how algorithmic bias works—not in theory, but in practice.

The Emerging Standards

The legal profession is beginning to develop standards for AI use. The ABA’s Resolution 604 (2024) on Generative AI emphasizes the duty of technological competence. Several jurisdictions now require disclosure of AI assistance in court filings.

For arbitrators, the standards are less formal but no less important:

Due Diligence: Understand the capabilities and limitations of any AI tool you use.

Verification: Never rely solely on AI-generated legal research or analysis. Verify citations, check primary sources, and confirm jurisdictional applicability.

Transparency: Consider whether AI assistance should be disclosed to the parties, particularly if it substantially shapes your analysis.

Independence: Ensure AI use enhances rather than replaces independent judgment. The arbitrator’s role is irreducibly human: weighing credibility, assessing context, applying judgment to ambiguous facts.

Conclusion: Discernment in the Algorithmic Age

AI will not replace arbitrators. But arbitrators who understand how models bend reality will think more clearly inside complexity than those who assume the machine is neutral.

In the end, bias isn’t the enemy. The unseen is the enemy.

Once you can see the tilt, you can steer around it.
Once you can name the pattern, you can break it.
Once you understand the worldview of the tool, you can decide when to trust it and when to hold the pen yourself.

The algorithmic age doesn’t demand perfection. It demands discernment.

And discernment, in its oldest sense, means this: the ability to tell one thing from another, to see the contours that others miss, and to navigate complexity with a mind awake to its own shadows.

That is the real skill of this era.
And it belongs to anyone willing to look closely enough.

References and Further Reading

Foundational Research

1. Bommasani, R., et al. (2021). “On the Opportunities and Risks of Foundation Models.” Stanford HAI Center for Research on Foundation Models. [Comprehensive analysis of bias in large language models]

2. Bender, E., et al. (2021). “On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?” FAccT ‘21. [Seminal paper on data bias and environmental/social costs]

3. Perez, E., et al. (2022). “Discovering Language Model Behaviors with Model-Written Evaluations.” OpenAI Research. [Demonstrates prompt sensitivity and interaction bias]

4. Maynez, J., et al. (2020). “On Faithfulness and Factuality in Abstractive Summarization.” ACL 2020. [Critical for understanding hallucination patterns]

5. Welleck, S., et al. (2023). “Generating Sequences by Learning to Self-Correct.” ICLR 2023. [Evidence for adversarial prompting effectiveness]

Legal and Ethical Frameworks

6. American Bar Association (2024). “Resolution 604: Generative AI and the Legal Profession.” [Establishes duty of technological competence]

7. Legg, M., & Song, J. (2024). “AI in International Arbitration: Promise and Peril.” Journal of International Arbitration, 41(2). [Specific to arbitration context]

8. Whittlestone, J., et al. (2019). “The Role and Limits of Principles in AI Ethics.” Journal of AI Ethics. [Philosophical foundations]

Practical Guides

9. Anthropic (2024). “Constitutional AI: Harmlessness from AI Feedback.” Anthropic Research. [Understanding design bias in Claude models]

10. OpenAI (2023). “GPT-4 Technical Report.” OpenAI Research. [Transparency about model capabilities and limitations]

11. Mollick, E. (2024). “Co-Intelligence: Living and Working with AI.” Portfolio/Penguin. [Accessible introduction to practical AI use]

Specialized Resources for Arbitrators

12. Malek, B., et al. (2024). "SVAMC Guidelines on the Use of Artificial Intelligence in Arbitration." Silicon Valley Arbitration & Mediation Center. [First comprehensive guidelines for AI use in arbitration practice]

13. International Chamber of Commerce (2024). “Guidance Note on the Use of Artificial Intelligence in ICC Arbitration.” [Practice-specific guidance]

14. Global Arbitration Review (2024). “AI and Arbitration: A Practical Guide.” [Case studies and practitioner perspectives]

15. UNCITRAL (2023). “Technical Notes on Online Dispute Resolution.” [Framework applicable to AI-assisted processes]

Model-Specific Documentation

15. Claude Documentation: https://docs.anthropic.com

16. OpenAI Documentation: https://platform.openai.com/docs

17. Google AI Documentation: https://ai.google.dev

18. Meta Llama Documentation: https://llama.meta.com

Ongoing Resources

19. Stanford HAI: https://hai.stanford.edu [Regular updates on AI bias research]

20. Partnership on AI: https://partnershiponai.org [Cross-organizational guidance]

21. AI & Law Forum: Various university law schools maintain active research programs

Note to Readers: This article reflects the state of AI technology and arbitration practice as of January 2026. Both fields are evolving rapidly. For current information on specific models or practices, consult the documentation and research resources listed above.

This article is researched and illustrated with AI tools (ChatGPT, Claude and Midjourney)

This is part of TechCred’s ongoing series on practical AI literacy for legal professionals. For questions or to suggest future topics, reach out through Substack.

Arbitration at the Forefront of the AI Era

Generative AI is reshaping how we think about evidence. Synthetic media—whether video, audio, or documents—has reached a level of realism that raises important questions about authentication. In litigation, these developments are sometimes framed as a crisis. In arbitration, they can be viewed as an opportunity. Arbitration’s hallmark flexibility makes it uniquely well-suited to adapt quickly, refine safeguards, and build trust in a digital era.

A widely reported 2024 case in Hong Kong, where deepfake video conferencing tricked a multinational firm into transferring millions, illustrates the technology’s power. Yet it also highlights arbitration’s advantage: unlike courts constrained by legislation, arbitral tribunals can experiment with innovative tools and protocols, positioning arbitration as a leader in shaping global standards for evidence integrity.

Arbitration’s Adaptability as an Asset

Arbitration has always thrived by combining flexibility with fairness. Party autonomy under the UNCITRAL Model Law allows procedures to be tailored to the dispute, while institutions like the ICC and LCIA set broad but adaptable frameworks. That same flexibility can be used to integrate new safeguards for synthetic evidence.

While some see vulnerability, others see agility. Courts must wait for legislatures to update statutory rules. Arbitral tribunals, by contrast, can respond immediately—through case management orders, disclosure protocols, and expert appointments. A 2025 Queen Mary/White & Case survey confirmed that arbitrators already anticipate a growing role for AI in proceedings, which means readiness is top of mind across the field.

Authentication as a Constructive Challenge

In the depths of a dispute, where light is scarce and appearances can mislead, deepfakes raise two key questions: how to authenticate digital evidence, and how to preserve procedural fairness.

• Authentication: Tools for forensic verification are improving rapidly, from metadata analysis to AI-driven detection software. Tribunals can require these tools as a matter of procedure, turning uncertainty into a structured evidentiary process.

• Procedural fairness: Far from undermining due process, these challenges give tribunals the chance to refine practices like cross-examination, disclosure, and expert testimony. By doing so, arbitration can offer a model of how justice adapts responsibly to technology.

Rather than assuming risks will overwhelm proceedings, these questions become opportunities to demonstrate resilience and creativity.

Practical Pathways for Tribunals

Arbitration can stay ahead of synthetic evidence by integrating solutions that are already emerging:

1. Metadata and Verification
   Require parties to provide metadata with digital submissions and enable forensic screening. The IBA Rules on Evidence already allow tribunals to assess probative value and exclude unreliable materials.

2. Appointment of Neutral Experts
   ICC and UNCITRAL frameworks empower tribunals to appoint experts. Applying this to AI forensics ensures impartial analysis of contested evidence.

3. Disclosure Protocols
   Tribunals can request early disclosure of whether AI tools were used in preparing evidence. Transparency prevents disputes from being derailed by surprise allegations.

4. Immutable Record-Keeping
   Secure timestamping and blockchain-based chains of custody can reinforce confidence in digital submissions, offering parties reassurance that evidence cannot be altered mid-proceeding.

5. Institutional Guidance
   Initiatives within ICC, CIArb, and SVAMC are exploring how AI intersects with arbitration. Tribunals can draw on this momentum and adopt new recommendations proactively at the case-management stage.

Each of these measures is less about guarding against disaster and more about reinforcing the credibility of arbitration as forward-thinking and robust.

Conclusion: Versatility Through Adaptability

Deepfake and synthetic-evidence threats challenge arbitration’s resilience. In navigating these deep waters, arbitrators can chart the course, showing how fairness and trust endure even in the darkest depths.

Synthetic evidence is best seen not as a crisis but as a proving ground for arbitration’s adaptability. With proactive integration of technology and updated institutional guidance, arbitral tribunals can turn what some see as a threat into an advantage.

The lesson is not that arbitration is vulnerable, but that it is versatile. As AI reshapes the evidentiary landscape, arbitration can prove again why it remains the preferred forum for complex, cross-border disputes: flexible, trusted, and ready to adapt.

Further Reading

• Financial Times, “Hong Kong Engineers Fooled in $25m Deepfake Video Scam” (2024) – A real-world example of how synthetic media is reshaping trust in digital communications.

• Berkeley Technology Law Journal, “Synthetic Evidence in Litigation: Authentication and Adaptation” (2024) – Academic discussion on how courts and tribunals can adapt evidentiary practices.

• ICC, Task Force on the Use of Information Technology in International Arbitration – Ongoing institutional work exploring technology’s impact on arbitral procedure.

• CIArb, Guidelines on the Use of Technology in International Arbitration (2024) – Practical recommendations for tribunals managing digital evidence and AI-related issues.

• UNCITRAL, Notes on Online Dispute Resolution (2021) – Guidance on managing digital submissions and remote hearings.

• SVAMC, Commentary on AI in Arbitration (2024) – Early thought-leadership on disclosure and transparency when AI tools are used in evidence preparation.

• Queen Mary University of London & White & Case, 2025 International Arbitration Survey – Insights into how arbitrators and practitioners view AI’s role in current and future disputes.

TechCred Session II (Oct 17, 2025) :
AI, Autonomy & Evidence That Doesn’t Sleep

These same themes echoed across Session II of TechCred’s inaugural program, where leading voices in the field examined the evolving responsibilities of arbitrators in the face of synthetic evidence, AI-assisted reasoning, and decentralized legal actors.

The message was clear: arbitration is no longer reacting to technological change—it is defining the standards by which justice will operate in the digital era.

////

This article is written together with AIs.

In his latest article, Barry Appleton argues that Canada must confront a growing crisis of algorithmic governance. Our digital reality—what news reaches us, what opinions circulate, what voices are amplified—is increasingly determined by a handful of global platforms whose operations remain opaque, unaccountable, and largely unregulated.

Appleton points out that this is not just a matter of consumer protection or data privacy. It is a constitutional issue. When platforms act as arbiters of speech and access to information, they effectively perform a public function—without public oversight.

But what happens when a country’s legal framework is built for human decision-makers, and decisions are now being made by code?

Courts can subpoena a minister, cross-examine a witness, or compel disclosure—but they cannot cross-examine an algorithm. Yet algorithms shape perception itself. They decide which facts gain traction, which opinions appear balanced, and which truths are quietly buried beneath an ocean of noise.

True oversight, then, requires more than transparency; it demands comprehensibility. We must be able to understand not just what algorithms are doing, but why—and to ensure that their reasoning aligns with our democratic values.

As Appleton writes,

“Algorithmic systems have replaced the editorial judgments of humans, but not their responsibilities.”

The challenge is to bring law, technology, and ethics back into conversation—to make sure those responsibilities are not lost in translation between code and conscience.

The TechCred accreditation program exists for precisely this reason: to prepare lawyers, arbitrators, and policymakers for a world where code itself can be a decision-maker. Algorithmic oversight is not only a regulatory challenge—it is a matter of justice, transparency, and democracy.

Further Reading

Barry Appleton

Read Barry’s full article: Algorithmic Oversight: Canada Must Confront Platforms That Control What We See
(Originally published in National Post, October 7, 2025)

.

Subscribe now

Written together with AIs

If you had to describe the digital age in three words, you could do worse than these:
Fake identities. Fake information. Fake behaviour.

That’s how a recent paper by TechCred faculty member Dmytro Chumachenko and colleagues frames some of the new challenges created by generative AI.
Not threats in the sci-fi sense, but shifts — quiet, steady shifts — in how we recognize what’s genuine in a world increasingly built on code.

A fake identity might be a composite photo of a person who doesn’t exist.
Fake information might be text written by a machine that sounds perfectly human.
Fake behaviour might be a swarm of invisible bots giving the illusion of a public reaction that never happened.

If that sounds abstract, think of how often we already rely on unseen systems to make decisions — search engines, credit scoring, recommendations, even legal analysis. These systems don’t always deceive us, but they do mediate reality. The “three fakes” remind us that what we see and what we trust are now designed experiences. Recognizing that isn’t cynicism; it’s the beginning of digital literacy for an algorithmic world.

Rather than seeing this as a crisis, we might see it as an invitation:
to rebuild trust with intention, to design systems that value transparency, and to remember that judgment — real human judgment — is still the most advanced technology we have.

Maybe that’s the work ahead.
And maybe that’s what TechCred has always been about.

REGISTER NOW

Read the full Balsillie Paper: Artificial Intelligence and New Threat Vectors: Using Scenario Planning for Trend Forecasting

.

This piece was written together with AIs.

.

Governance once meant parliaments, ministries, and courts. Today, it’s also written in code — in how data moves, how models learn, and how rules are automated. Each time a policymaker uses an AI tool or a tribunal evaluates digital evidence, a question surfaces: who governs the governor?

Law now lives inside architecture. Every algorithm is a policy choice, every dataset a jurisdictional line. As systems cross borders, their logic does too. Institutions that once defined ethical norms—universities, courts, regulators—now depend on technologies governed elsewhere. This isn’t conspiracy; it’s inertia. We built global infrastructure before we built global legitimacy.

Dr. Dmytro Chumachenko’s research on academic integrity offers a model for this moment. He maps five layers of governance—international, national, institutional, professional, and individual—showing how ethics can survive across borders through coordination, not control.

Replace academic integrity with algorithmic integrity and the principle still stands: legitimacy in complex systems is not imposed, it’s negotiated.

The problem is that the world’s digital scaffolding remains largely private and opaque. We regulate what we see—the apps, the outputs—but not the layers beneath. Cloud contracts, inference engines, and content filters now act as meta-laws, shaping what can be known or believed. Without oversight, automation risks becoming the illusion of governance. Speed isn’t legitimacy.

Barry Appleton calls this the new form of sovereignty—rule by computation. Chumachenko reminds us that integrity is never held by one level alone. It depends on constant dialogue between global principles and local realities. True digital sovereignty is not the power to exclude, but the ability to participate responsibly in systems that include everyone.

If the nineteenth century asked who owns the land, and the twentieth who owns the oil, the twenty-first asks who owns the logic. The answer will determine whose values are encoded into the systems that govern daily life. Governance must evolve from regulation to architecture—from writing rules to designing systems that make fairness the default.

That is the new test of legitimacy.

This article was inspired by recently published works by Prof. Barry Appleton and Dr. Dmytro Chumachenko

Read more

Barry Appleton: AI and cloud infrastructure is the railway of the future — why isn’t Canada building it?

Published September 30, 2025 National Post

Artem Artyukhov & Dmytro Chumachenko: International organizations in promoting academic integrity in a globalized world: Ukrainian view

Published July 9, 2025 Geopolitics under Globalization Journal Volume 6

/// Written together with AIs.

When we think of sovereignty, we picture borders, flags, and parliaments. But in the algorithmic age, sovereignty is increasingly exercised through contracts, cloud servers, and code.

Every time a government official logs into a cloud-based platform, or a court relies on an AI system to evaluate evidence, a question emerges: who really governs that decision?

The Invisible Border Crossings

Most people believe their digital government services are “local”—that a tax return filed in Paris, São Paulo, or Nairobi is managed by their own institutions. Yet the truth is more complicated. Data often flows through infrastructure designed abroad, stored on servers subject to foreign law, and processed by algorithms optimized for someone else’s interests.

The border crossings are silent. No passport check. No foreign anthem. Just a login, a click, a query—and suddenly, jurisdiction shifts.

The Rise of the Algorithmic Empire

Unlike empires of the past, today’s power is not measured in ports, railways, or armies. It’s built on control of digital bottlenecks:

• Cloud platforms hosting government records.

• AI engines sorting visa applications or scanning health data.

• Recommender systems shaping what populations see and believe.

• Digital payment rails that govern the movement of money.

Together, these systems create what Professor Barry Appleton calls an algorithmic empire: governance enforced through technical standards and computational architecture, rather than soldiers or treaties.

Citizens experience it as convenience. Lawyers describe it as “jurisdiction without borders.” But in reality, it’s a quiet outsourcing of sovereignty.

Dependency Disguised as Modernization

When any government relies on Microsoft 365, AWS, or a foreign-designed AI engine, it doesn’t simply adopt a tool—it accepts another country’s legal framework and business priorities. What looks like modernization may actually be systemic dependency.

The risks are not theoretical. Laws like the U.S. CLOUD Act allow foreign authorities to access data even when it never leaves national soil. Procurement orders and software updates become the modern equivalents of treaties and territorial concessions.

Why This Matters Everywhere

This is not only a Canadian story. The same dilemmas face European regulators navigating U.S. tech dominance, African states weighing Chinese cloud contracts, and Latin American governments drafting AI rules under the shadow of extraterritorial norms.

Arbitrators, lawyers, and policymakers cannot treat these as narrow IT issues—they go to the heart of legitimacy, jurisdiction, and enforceability in a digital world.

A Canadian Case Study

For a deeper dive into how these issues play out in practice, international trade lawyer, Professor Barry Appleton has written a powerful essay on Canada’s experience of ceding sovereignty to America’s algorithmic empire. It offers a sharp case study of what’s at stake when governments outsource critical digital infrastructure.

👉 Read the Canadian perspective here

_____

This article was written together with AIs.

.

Arbitration’s Role in the Digital Trade Era

Imagine you’re arbitrating a cross-border dispute when an AI-driven customs tool sparks debate: who controls the data behind the decision? For decades, trade disputes focused on tariffs on steel, soybeans, or cars. Today, the contested flows are digital—data, algorithms, and AI systems that underpin global commerce. For arbitrators and mediators, this shift is not abstract. It is already reshaping the disputes you will see, and it invites you to help define fairness in the digital era.

Just as on a golf course, where two nations plant their flags on the same green, the challenge lies in ensuring the contest is played by rules everyone accepts.

The Rise of Digital Trade Clauses

Modern trade agreements already reach far beyond goods. The USMCA, the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP), and the Digital Economy Partnership Agreement (DEPA) all include digital trade chapters. These provisions cover everything from cross-border data flows and source code disclosure to algorithmic transparency.

At the multilateral level, the World Trade Organization (WTO) moratorium on customs duties on electronic transmissions, in place since 1998, is under growing scrutiny as some members push to let it lapse. The debate captures a historic pivot: whether “transmissions” of data should be treated like goods, taxed like goods, and litigated like goods.

For arbitrators, this means that future disputes may ask whether:

• a restriction on cross-border data violates a digital trade commitment,

• an algorithmic tool at a digital “border” acted fairly, or

• national security justifications for restricting data flows are proportionate.

In other words, the game is moving from tariffs to transmissions — and arbitrators are being asked to keep score on a brand-new course.

What Digital Sovereignty Means for Disputes

At the same time, digital sovereignty has become a policy watchword. The EU has led with the AI Act and GDPR, setting strict standards for high-risk AI and data protection. The U.S., through measures like the CLOUD Act, emphasizes law enforcement access to data across borders. Many states are moving to assert their own frameworks, rather than import others’.

For dispute resolution, this raises the problem of standards equivalence: how to reconcile one country’s AI and data rules with another’s. Arbitrators may find themselves comparing regimes the way golfers compare scorecards—making sure the game is measured fairly even if the strokes were played under different local rules.

Without agreed rules, every nation wants to play its own course. But in global trade, the fairways inevitably overlap.

The Arbitrator’s Role in the Digital Era

So what does this mean in practice? Expect disputes that blend law and technology:

• Algorithmic Transparency: Was an AI system used in trade or customs explainable and unbiased? Arbitrators may need to interrogate its design much as they would test a witness’s credibility.

• Cross-Border Standards: Do measures on data or AI meet obligations under agreements like CPTPP or USMCA?

• Economic Security: How should tribunals weigh national security justifications against commitments to free digital trade?

These are new questions, but not new skills. Arbitrators have always been guardians of fairness — referees ensuring the match is played within the rules, even when those rules are in flux.

How to Prepare

Preparation means building fluency in both law and technology. Resources already exist:

• SVAMC’s Guidelines on AI in Arbitration (2024) provide protocols for handling AI-generated evidence.

• Initiatives like TechCred offer training for arbitrators seeking literacy in digital systems.

• The Queen Mary/White & Case International Arbitration Survey (2025) shows that over 70% of practitioners expect AI to reshape dispute resolution.

Arbitration’s great strength is adaptability. And in a contest where the landscape itself is shifting, adaptability is what keeps the fairway open.

Conclusion: Your Seat at the Digital Table

Digital trade clauses aren’t just policy—they are becoming the terrain of dispute resolution. By mastering the interplay of law, technology, and sovereignty, arbitrators can ensure transparency and fairness in this new frontier.

The digital frontier is not a level course yet. Whether it becomes a fairway—or a minefield—depends on how disputes are judged. Arbitrators now have the chance to lead, ensuring the game is played fairly, no matter whose flag is on the green.

What ideas are you bringing to this frontier? Join the conversation in the comments—or at TechCred. We are all learning together.

Further Reading

• WTO, “E-Commerce Moratorium and Digital Trade Developments” (2024) – Overview of debates around taxing electronic transmissions.

• European Commission, “EU AI Act” (2024) – Regulations on high-risk AI, critical to digital trade contexts.

• U.S. Department of Commerce, “Digital Economy Report” (2023) – U.S. policy framing for digital trade and data flows.

• SVAMC, “Guidelines on AI in Arbitration” (2024) – Practical guidance for integrating AI into dispute resolution.

• Queen Mary/White & Case, “International Arbitration Survey” (2025) – Practitioner perspectives on AI and arbitration.

• Barry Appleton, “From Tariffs to Transmissions: Why the Trump Round Puts Digital Trade at the Centre of Global Negotiations” (Clause & Effect Blog, Aug. 21, 2025) – A timely commentary framing digital trade as constitutional in scope.

Barry Appleton

From Tariffs to Transmissions:

From Tariffs to Transmissions: Why the Trump Round Puts Digital Trade at the Centre of Global Negotiations…

Read more

10 months ago · Barry Appleton

.///

This article was written together with AIs.

.

.

Barry Appleton describes the infrastructure trap as the moment when governments and regulators find themselves constrained because essential systems—railways, pipelines, or today’s digital platforms—are privately controlled. Once you depend on them, your choices narrow.

Digital giants like Facebook and Google rule as if their platforms were private planets. They are not states, yet they govern commerce, finance, speech, and even justice with a reach greater than many countries. Their rules are binding not because we voted for them, but because daily life depends on them.

For the international dispute resolution community, this raises familiar questions—neutrality, fairness, enforceability—but in a new, digitally driven context.

Digital Platforms as Private Adjudicators

Many of these “private planets” already run courts of their own.

• E-commerce platforms adjudicate refund requests.

• Financial technology providers resolve investor disputes in-app.

• Social media companies use AI-moderated systems or ratings arbitration to decide speech conflicts.

These internal processes are fast, scalable, and efficient. Yet from an arbitrator’s perspective, they highlight old concerns in new forms:

• Procedural standards vary widely. Each platform writes its own rules, with no UNCITRAL or ICC to harmonize them.

• Transparency is limited. Many systems resemble closed chambers, with little chance to test credibility.

• Neutrality is compromised. The sovereign of the platform is often both party and judge.

For arbitrators, the lesson is clear: private adjudication is not a novelty. It is already here, and it is reshaping expectations of how disputes are resolved.

The Expanding Role of Online Dispute Resolution (ODR)

Governments and international organizations are also experimenting with scalable online dispute resolution (ODR) systems:

• UNCTAD has piloted blockchain-enabled ODR projects in Asia.

• Several European regulators are standardizing consumer ODR.

• Australia’s national platform resolves huge volumes of consumer cases online before escalation.

To experienced arbitrators, much of this will look familiar. ODR borrows from arbitration and mediation frameworks but retools them for speed and volume. And just as cross-border trade once required arbitral innovation, digital disputes rarely stay confined to one jurisdiction—making international perspectives essential.

In other words: disputes are no longer just moving between states. Increasingly, they travel across these private planets.

Implications for International Practitioners

For arbitrators and mediators, the rise of digital platforms as infrastructure creates both challenges and opportunities:

• Neutrality: How can fairness be guaranteed when the forum is designed by the very entity with the greatest stake in the outcome? Arbitrators can adapt their long experience of checking conflicts of interest and party control to these new digital forums.

• Due process at scale: What safeguards exist when disputes are resolved by algorithm, not advocate? Arbitrators can help design standards that balance speed with procedural fairness.

• Predictability: Without harmonization, parties risk very different processes depending on which platform their dispute lands on. Arbitrators’ comparative skills make them natural translators between systems, ensuring consistency across borders.

• Engagement: Platforms are already shaping their own rules. By engaging early, arbitrators can bring legitimacy and due process into systems that might otherwise tilt entirely toward efficiency.

This is not a question of replacing courts or institutions. It is about acknowledging that platform sovereignty now exists alongside state sovereignty—and that ignoring it leaves practitioners one step behind.

Why This Matters for Arbitration

Platforms may seem new, but the issues they raise—who controls the forum, who sets the rules, how fairness is safeguarded—are familiar territory to arbitrators. The infrastructure trap is less a departure than an extension of principles long defended in practice.

Imagine if one party in your case built the hearing room, owned the filing system, and employed the stenographer. That is how today’s platforms operate. For arbitrators, the infrastructure trap is not just a policy debate. It is a professional challenge—and an opportunity to ensure fairness does not get lost in translation as justice goes digital.

Conclusion

Digital platforms are not supplanting arbitral institutions, but they are becoming the terrain on which many disputes will first arise. They may not have constitutions or appellate courts, but their decisions shape commerce and daily life.

Arbitrators are not bystanders here. The very skills you have honed—balancing neutrality, safeguarding due process, reconciling clashing legal systems—are the ones most needed as digital forums expand. Private platforms may set the rails, but arbitrators can bring accountability, predictability, and legitimacy into their orbit.

The infrastructure trap reminds us that tomorrow’s disputes will be fought on private terrain. The opportunity now is to ensure those terrains remain fairways—open, navigable, and governed by the principles of justice that have always anchored arbitration.

Further reading

• UNCTAD, Blockchain and ODR Pilots in Asia (2023).
  Illustrates how international organizations are experimenting with scalable digital dispute resolution and what lessons might transfer to arbitration.

• European Commission, Consumer ODR in the EU (2023).
  Shows how regulators are building standardized ODR frameworks, offering models that could shape expectations for cross-border disputes.

• Australian Competition and Consumer Commission, ODR Platform Reports (2024).
  Provides a national case study of high-volume digital dispute resolution in practice—useful for arbitrators thinking about scale and due process.

• Original article that inspired this piece: “WonderFi and the Surrender of Financial Sovereignty. Why Canada must block the Robinhood takeover of its digital rails.”

  Explores how private control of digital infrastructure can limit government sovereignty—an early warning of the “infrastructure trap.”

Barry Appleton

WonderFi and the Surrender of Financial Sovereignty

WonderFi and the Surrender of Financial Sovereignty…

Read more

a year ago · 2 likes · Barry Appleton

////

This article was written together with AIs.

.

Contracts have historically been agreements between humans or their legal proxies, such as corporations. But what happens when one party is an autonomous AI system, acting independently of human oversight?

Consider a scenario: an AI agent, programmed to negotiate supply contracts within defined parameters, finalizes a deal at 2:14 a.m. The human company discovers it is bound by terms no manager reviewed. When a dispute arises—over pricing, delivery, or compliance—how is it resolved, and where? For seasoned arbitration practitioners, this question signals a new frontier in dispute resolution.

International arbitration has long adapted to commercial shifts, from cross-border trade to digital signatures and smart contracts. Yet, an autonomous AI contracting party raises a fundamental issue: who—or what—constitutes the “party” in a dispute?

Legal Personhood and Accountability

In the U.S., corporate law treats entities as legal “persons,” suggesting tribunals might view AI agents as tools of their human or corporate principals. For example, in the 2018 Uber pricing algorithm controversy, the algorithm’s actions were attributed to the company, not the AI itself. But what if an AI exceeds its programmed mandate, creating terms its principal didn’t anticipate? In Europe, discussions around “electronic personhood” (e.g., the EU Parliament’s 2017 resolution on robotics) might prompt tribunals to consider limited standing for AI systems, though this could blur lines of human accountability. In Asia, tech-forward arbitral hubs like Singapore’s SIAC or Hong Kong’s HKIAC may lead in crafting rules for AI-driven disputes, building on their experience with e-commerce arbitration.

Procedural Challenges for Arbitrators

Arbitration’s flexibility positions it to address these disputes before national courts establish rigid precedents. Practitioners will face several practical questions:

• Evidence and transparency: How should tribunals evaluate an AI’s decision-making? Requiring disclosure of algorithms or training data could clash with proprietary interests, yet opacity risks unfair outcomes.

• Party representation: If an AI negotiates independently, can it be “represented” in proceedings, or must counsel interpret its actions? This could redefine traditional notions of advocacy.

• Contractual intent: When two AIs negotiate without human oversight, how do arbitrators assess the “meeting of the minds” central to contract law?

• Ethical considerations: What if one party’s AI, leveraging superior processing power, creates an imbalance in negotiations? Arbitrators may need to develop standards for auditable AI decision logs to ensure fairness.

These challenges highlight arbitration’s unique role as a testing ground for AI-related disputes. Institutions like the ICC or LCIA could adapt existing frameworks, such as expedited procedures, to handle machine-made agreements. For instance, UNCITRAL’s 2016 Notes on Organising Arbitral Proceedings offer guidance on structuring hearings involving novel evidence, which could apply to algorithmic outputs.A New Arbitration Landscape

The rise of AI parties raises deeper questions about the nature of legal responsibility. If both sides’ AIs negotiate terms neither principal foresaw, who bears liability? What if an AI’s “decisions” are the heart of the dispute, and one side is represented by counsel while the other relies on a counsel-assisted algorithm? These scenarios push arbitrators to rethink foundational concepts like intent, agency, and fairness.

Arbitration’s strength lies in its ability to evolve case by case, across jurisdictions, without waiting for fragmented national laws to coalesce. Yet, this adaptability will be tested as tribunals confront disputes where one “party” is not human. The real challenge is not whether arbitration can handle such cases—it has the tools—but whether practitioners can navigate the intersection of law, technology, and ethics to deliver just outcomes.

This new reality invites arbitrators to anticipate how autonomous systems will reshape commerce and dispute resolution. As AI increasingly mediates agreements, arbitration may become the first forum to define the legal contours of a world where the other party at the table is no longer human.

__________________________________________

Further Reading

Foundational depth:

• Born, Gary. International Commercial Arbitration (3rd ed., 2021)

• Lew, Mistelis & Kröll. Comparative International Commercial Arbitration (2003)

Procedural anchors:

• UNCITRAL Notes on Organising Arbitral Proceedings (rev. 2016)

Accessible synthesis:

• Redfern & Hunter. Law and Practice of International Commercial Arbitration (7th ed., 2023)

Historical foresight:

• Kaufmann-Kohler & Schultz. Online Dispute Resolution: Challenges for Contemporary Justice (2004)

Tech Specific Insights:

• SIAC Rules (2021), Section on Technology and Dispute Resolution

Current debates:

• Wolfe, Stephen. “AI and International Arbitration: Opportunities and Risks” (Journal of Int’l Arb., 2022)

///

This article was written and edited with AIs.

.