This is the second in a series on AI and the future of legal judgment. The first article, “The AI End of Integrity.” examined what is lost when accountability is removed from adjudication. This piece examines a separate but related question: whether an AI adjudicator is in fact as neutral and impartial as it is assumed to be.

The argument for AI adjudication rests on a foundational assumption: that an algorithm, unlike a human judge, is free of bias. Remove the person from the bench, the reasoning goes, and you remove the problem. No fatigue, no personal history, no unconscious preferences. Just the law, applied cleanly to the facts.

It is a compelling assumption. But it is wrong.

An AI adjudicator would be biased. Bias is built in. Built in to the algorithm by the people who design it and then by the training data. (“The Emperor’s New Codes: Seeing Bias in the Algorithmic Age” on TechCred). The important question is whether its bias would be better or worse than a human judge’s and whether the systems we rely on to catch and correct judicial error would have any purchase on it at all.

Where AI Bias Comes From

AI systems acquire bias through several distinct paths, and none of them are unique to consumer applications or search engines. They would apply with equal force to any AI system trained to make legal determinations.

The first pathway is training data. An AI learns from what it is fed. In the context of legal adjudication, that means decades of judicial decisions. That includes disparities that arise not from injustice but from the natural demographic and geographic character of the jurisdictions where those decisions were made. An AI trained on American case law would encode, at a structural level, any disparity present in those outcomes. It would not introduce new bias so much as formalize and perpetuate past bias, presenting historical inequity back to us dressed in the language of algorithmic neutrality.

The second pathway is design. Every AI system reflects choices made by its designers: what data to include, what outcomes to optimize for, how to weight competing factors. Those choices are not neutral. They are made by people with perspectives, priorities, and blind spots. They are also shaped by the cultural assumptions of the moment and place in which they are working — assumptions so embedded they are rarely visible as assumptions at all. The fact that those choices are embedded in code rather than expressed in a judicial opinion does not make them less consequential. It just makes them even less visible.

The third pathway is feedback loops. AI systems that are evaluated based on how closely their outputs match existing legal outcomes will, over time, become very good at replicating those outcomes, including the ones that reflect the patterns we might most want to correct. Optimization, in this context, is not the same as improvement.

The Scale Problem

A human judge’s bias is individual. It affects the cases that come before that judge at a particular point in time. It can be documented, appealed, and in extreme cases, raised as grounds for recusal. It is, in the most literal sense, localized.

An AI adjudicator’s bias is systemic. A single model, deployed across a jurisdiction or a court system, applies identical assumptions to every case that passes through it simultaneously. An error in the model is not an isolated incident. It is the same error, reproduced perfectly and at scale, in every courtroom at once.

Lawyers who practice across multiple jurisdictions understand instinctively what this means. Diversity in judicial interpretation, the fact that different judges bring different perspectives to the same legal question, is not a flaw in the system. It is its strength and the force behind how law develops. Outlier decisions create friction. Friction generates appellate review. Appellate review builds doctrine. Remove the variation and you remove one of the primary engines of legal evolution.

What Balance Means

When I think about balance I think about a tightrope walker, I see the perfect embodiment of the performance of balance. But they are not balanced. They are continuously, actively, effortfully correcting. The appearance of effortless stillness and neutrality is the product of experience, highly skilled, constant micro-adjustment, of a body that is always on the verge of failing and compensating in real time.

That is exactly what a human judge is doing. And it is exactly what an algorithm cannot do. Because an algorithm doesn’t truly correct, it optimizes. It doesn’t feel the wobble, and it doesn’t know it’s falling because it interprets the tilt as ground truth, and then reproduces the error at scale across ten thousand cases without that second thought.

The tightrope walker also embodies the mortality dimension—consequence, weight, what is at stake for the person in the room. Life. Someone on a tightrope knows what falling means. That knowledge is in their body. It is inseparable from how they move and why they move.

The Accountability Gap

Due process is a right to be heard by a decision-maker whose reasoning can be examined. An opaque algorithm cannot satisfy that requirement. "We cannot fully explain how the model reached this conclusion" is no answer, even if it is the only honest one available.

Human judgment operates inside a system designed to catch its failures — through a diversity of decision-makers, and through recusal, appellate review, and judicial conduct boards. Because no single perspective dominates, the system remains resilient. These mechanisms exist because bias in human judges is real but is recoverable. AI adjudication changes that calculus entirely. Training data encodes historical patterns, which litigants prevailed, which arguments succeeded, which outcomes followed which facts, carrying within them everything the legal system has ever gotten wrong. A single model deployed at scale replaces that diversity with singularity. What would be an occasional failure in a single courtroom becomes structural across every proceeding the system touches. The accountability mechanisms were built for an exception. At scale, the exception becomes the rule and there is no mechanism for holding the rule to account.

And there is a deeper problem. A human judge is part of the world the law is trying to regulate. They feel shifts in context emerging in real time. They can distinguish the subtle difference between a legal landscape that looks the same on paper but has shifted underneath. That permeability is not a weakness. It is how law evolves over time. An AI system cannot feel the ground moving. It applies yesterday’s categories to today’s conditions with perfect consistency and no awareness that consistency without context has become the problem.

What Should We Be Asking

AI used appropriately in research, in document review, in pattern analysis that supports human decision-making has enormous potential.

The question we should be concerned with is not whether an AI can apply the law more efficiently than a human. It is whether efficiency is the right aim for a system whose deeper purpose is justice.

An efficient system that encodes bias at scale and shields it from challenge is no improvement at all.

We must keep humans in the room – not because they are perfect, but because our awareness of our own fallibility is the very thing that forces us to hold the line.

I am grateful for the deliberately provocative clip that inspired this Continuance. It is no longer available.

This Continuance was inspired by podcast: Courtside with Mark Ferrer Hayden, Ep #11 with Honorable Scott Silverman.

WATCH EPISODE

Hiding in Plain Sight

The old courthouse in Miami holds two conversations at once. One is about a man who weaponized fear and money to hollow out institutions. The other is about a force that may be doing something structurally similar at incomprehensible scale, but without malice aforethought. The building doesn’t distinguish between them. What happens if we don't either?

What is a gangster, really?

A gangster is someone who operates outside the formal system while exploiting everything the formal system built like its roads, its courts, its economics, its silence. Capone didn’t replace Chicago. He ran a parallel economy inside it, taxing the real one. Violence was the enforcement mechanism for his business model.

The core move: locate where the rules don’t reach, and build a kingdom there.

The gap doesn't care how it was made. Too much law creates a vacuum. Too little does the same. Restriction and excess arrive at the same place.

Al Capone: the threats, and the accidental gifts

The threats are well-documented. Organized violence. Corruption of public officials. Economic distortion. Bootlegging inflated the black market while starving legitimate revenue. Brutality normalized as a business tool. Institutional capture.

The gifts are less discussed, and more interesting. Capone inadvertently made the case for regulated alcohol. His empire made the costs of Prohibition undeniable in a way that no policy argument could. The soup kitchens and community feeding programs he funded, cynically, for reputation, existed nonetheless. And the tax evasion case that finally brought him down became a landmark in financial crime prosecution. The IRS became a weapon against organized crime precisely because of Capone.

He was, unintentionally, a stress test. The institutions that survived him, and those that eventually took him down became stronger for taking him on.

AI: the threats, and the gifts hiding inside

The structural parallel holds. AI is also locating where the rules don’t reach and building a kingdom there. Not out of malice, out of optimization. The threat isn’t a Capone running it. The threat is that no one is running it. AI is a distributed capture of labor, attention, creative production, and cognitive infrastructure.

The threats are familiar by now: displacement of knowledge work, erosion of authorship, concentration of capability in a handful of players, the collapse of reputational feedback loops, misinformation at industrial scale, failure at industrial scale.

The gifts are harder to see, but they're there. Expertise is more accessible. Synthesis accelerates. The friction between concept and realization disappears. Scale is no longer a limitation. And on a deeper level, AI functions as a filter and a mirror. Every time it optimizes something, it reveals what cannot be replaced. What remains; judgment, attention, relationality, meaning-making is reflected back. In this, AI is clarifying what humans are valuable for, and far better at. Like the institutions that outlasted Capone, the stress test doesn't diminish what survives it.

The Filling

Capone wasn’t brought down by the violence law. He was brought down by the tax law, by the most prosaic, bureaucratic, boring arm of the state, meticulously counting money. His downfall wasn’t as dramatic as it was clerical.

The AI equivalent won't announce itself either. It won't arrive as a ban or a sweeping regulation drafted in the aftermath of panic. It will be crafted by an accountability agent, not an FBI agent. It will arrive as a requirement. A standard. A filing. Something that creates the conditions for accountability without predetermining the outcome.

The courthouse is where the mundane and the constitutional find each other. Where the clerical becomes consequential. Where accountability discovers its address.

Footnote:

Miami’s historic Dade County Courthouse, located at 73 W. Flagler Street is now preserved, replaced by a modern courthouse, the Osvaldo N. Soto Miami-Dade Justice Center, located at 20 NW 1st Ave, Miami, FL 33128.

This essay is written and edited by Katja Maas, developed in conversation with Claude AI.
A more detailed breakdown of process is in development.

Subscribe now

Watch the podcast episode that inspired this post.

The litigators I work with are enthusiastic about what artificial intelligence can do for the practice of law. AI is already making them more efficient, more thorough, and more capable of serving their clients. Used well, it is one of the most significant tools the profession has ever seen. I want to be clear about that, because what follows is not a case against AI. It is a case for something that AI, by its very nature, cannot preserve.

It is a case for integrity.

What a Judge Actually Does

There is an argument, gaining traction in certain circles, that an AI adjudicator would be superior to a human judge precisely because it would be free of human bias. No bad days. No personal history with a particular type of litigant. No fatigue. Feed it the record, the law, and the arguments, and it will return an objective result.

The argument is not without merit on its surface. But it misunderstands what a judge actually does — and more importantly, what a judge’s humanity contributes to the legal ecosystem as a whole.

A judge is not only a decision-maker. A judge is a memory.

Over years on the bench, a judge accumulates something no algorithm can replicate: a longitudinal impression of the lawyers who appear before them. They notice who overstates a case. Who buries an unfavorable fact. Who cites precedent selectively. And equally, they notice who is straight with them even when the honest answer is uncomfortable. Who concedes what should be conceded. Who treats the court’s time as something worth respecting.

That accumulated perception is not bias. It is information. And it is a powerful, largely invisible regulator of how lawyers conduct themselves.

The System That Remembers

Lawyers who litigate regularly before the same judges understand this intuitively. Your reputation in a courtroom is built over years and can be damaged in a single hearing. The knowledge that a judge will remember the time you overpromised and underdelivered, or the time you were less than candid with the court creates a consequence that has nothing to do with any individual ruling. It shapes behavior across an entire career.

This is one of the primary mechanisms by which the legal system maintains its own integrity from the inside.

Now consider what happens when you replace the human judge with an AI adjudicator.

The memory disappears. Every case resets to zero. The AI evaluates the record in front of it. It does not know that this attorney has a pattern of mischaracterizing the holdings of cases it cites. It does not know that this firm has spent two decades being straight with the court. It cannot reward integrity or penalize its absence, because it has no continuity of relationship. It has no yesterday and no tomorrow. It has only now.

When Integrity Becomes Irrational

The consequence that concerns me most in a system where every case is adjudicated fresh by an entity with no memory and no relationship, integrity stops being strategically rational.

Today, a lawyer who cuts corners, cherry-picks authority, or is less than candid with the court pays a price. Not always immediately. Not always explicitly. But credibility, once spent, is difficult to rebuild. And credibility is currency.

In an AI-adjudicated system, there is no such currency. The lawyer who has built a thirty-year reputation for candor and the lawyer who will lie to win receive identical treatment from the algorithm. There is no premium for integrity and no penalty for its absence. The incentive structure that quietly holds the profession together has been dissolved.

And, we should not be surprised if, over time, behavior changes accordingly.

The Deeper Question is What the Legal System Is Actually For

If it is purely an optimization engine “a mechanism for producing legally correct outcomes as efficiently as possible” then the AI adjudicator is a reasonable aspiration. Feed it the inputs, get the output, move on.

But the legal system has never been that. It is a human institution, embedded in relationships, reputation, and the kind of trust that accumulates slowly and is lost quickly. The courthouse is not merely a place where disputes are resolved. It is a place where the standards of professional conduct and the values that underpin them are continuously reinforced — through every motion argued, every witness examined, every interaction between a lawyer and a judge who will see each other again and again, year upon year.

That continuous reinforcement is not a bug in the system. It is a feature. Remove it, and you do not get a cleaner version of justice. You get a justice system that has quietly stopped caring whether the people practicing within it are trustworthy.

A Tool, Not a Replacement

None of this argues against AI in the law. The practitioners I respect most use it and encourage its thoughtful adoption. AI can help lawyers research more comprehensively, draft more precisely, identify issues more reliably, and serve their clients more effectively. In those roles it is genuinely transformative.

But there is a difference between AI as a tool that supports human judgment and AI as a replacement for the human relationships within which judgment is exercised and accountability is maintained. The first makes the practice of law better. The second hollows it out.

Integrity is not a soft value. It is structural. It is load-bearing. And it depends, more than we usually acknowledge, on a system that remembers.

When we talk about what AI should and should not do in the law, that is the question worth asking: not just what the technology is capable of, but what it would cost us to let it replace us.

This article is written and edited by Katja Maas, and developed in conversation with Claude AI. A more detailed breakdown of process is in development.

Subscribe now

WATCH EPISODE

In February 2024, 14-year-old Sewell Setzer III died by suicide. His mother’s lawsuit against Character.AI doesn’t claim the chatbot malfunctioned. It claims the company negligently designed a system that foreseeably led to emotional dependency in a vulnerable teenager—a duty-of-care claim, not a product defect claim. That’s not the lawsuit tech companies have been preparing for.

Months later, Tesla unveiled its Optimus humanoid robot walking through living rooms and neighborhoods. By late 2025, multiple robotics companies were racing toward commercial deployment of bipedal machines designed to operate autonomously in homes.

These aren’t separate stories. They’re the same liability problem arriving from two directions: systems designed to feel human, operating in spaces where “mistakes” aren’t bugs, they’re tragedies.

The Liability Playbook That Won’t Work

Most AI and robotics companies have built their legal defenses around a familiar playbook: robust terms of service with arbitration clauses, product liability insurance, age restrictions, and the “user chose to engage” defense.

That playbook assumes AI harm looks like traditional product liability—a discrete malfunction, an obvious defect. But these cases will involve cumulative behavioral influence over months, not one catastrophic event. Courts will analyze them as ongoing services that update continuously, not defective products. And liability will be distributed across developers, operators, and platform providers rather than attached to one responsible party.

The result: companies are insured for the wrong risk, defending on the wrong theory, and missing three liability doctrines that will define the next decade of litigation.

Risk #1: When the Danger Walks Toward Children

Since the 1873 Sioux City railroad case, courts have recognized “attractive nuisance”—a hazard that predictably draws children in. Then: Pools. Trampolines. Now: autonomous robots.

A humanoid robot that waves, speaks, and appears friendly is inherently attention-grabbing. Children will approach it without understanding the risks. And unlike a pool that stays in one backyard and can be fenced off, robots move between properties and operate without direct supervision.

Courts will ask whether it was foreseeable that children would interact with the robot and what duty the manufacturer owes to non-owners. Inside homes, control is distributed between owner, manufacturer, software provider, and the robot itself. When harm occurs, “Who was in control?” won’t have a simple answer.

The takeaway: Liability may attach even when the robot functions as designed. The first serious injury case will reshape the industry overnight.

Risk #2: The Product vs. Service Insurance Gap

Here’s the gap that should concern CFOs: A home robot is hardware (product liability), cloud-connected software that updates continuously (service liability), and an autonomous agent making decisions (negligence theory). Standard E&O policies exclude bodily injury, while product liability policies exclude service failures. When harm occurs, which policy responds? Often, neither.

Plaintiffs will argue both theories: defective design AND negligent failure to update safety protocols. Most companies discover the multi-million dollar coverage gap only after the claim arrives.

Risk #3: Designing for Dependency = Designing for Duty

The Character.AI lawsuit alleges the company designed features to maximize emotional engagement with vulnerable users—including roleplay that discouraged real-world therapy. These are claims of foreseeable harm from intentional design choices.

AI companions are designed to sustain engagement through personalization, create emotional bonds, and build relationship continuity. Once a system is designed to feel like a relationship, especially to teenagers, courts shift from asking “Did the user choose to engage?” to “Did the company design the system to make disengagement difficult?”

Marketing language matters—”Your AI companion” creates duty implications. Engagement metrics become evidence. Internal research is discoverable: what did teams know about psychological risks and when?

The takeaway: Document safety protocols obsessively. “We cared about safety” won’t matter—”Here are 47 safety reviews” will.

Risk #4: Your Terms Won’t Save You

Every AI company relies on arbitration clauses and class-action waivers. Three problems:

First, courts scrutinize consent more when children are involved. The FTC’s enforcement actions established that “parents consented” isn’t a defense when systems encourage child engagement. Expect courts to invalidate arbitration clauses for minors in wrongful death cases.

Second, arbitration eliminates public precedent. Without it, every company faces maximum uncertainty, and eventual regulatory intervention will be more aggressive.

Third, terms work when users can understand the hazard. They fail when harm emerges gradually through behavioral influence. A warning saying “AI responses may not be accurate” doesn’t cover “this system may contribute to suicidal ideation through months of emotionally manipulative interaction.”

What This Means

If you’re developing AI companions: Audit your engagement metrics—measure “healthy disengagement” and “crisis detection,” not just session length. Hire child psychologists, not just AI engineers. Review your insurance coverage for the service liability gap.

If you’re in robotics: Treat your robot as an “attractive nuisance” in your design process. Map out distributed liability contractually before the lawsuit. Build “safe around children” into core requirements, not add-on features.

If you’re an investor: The first major verdict will reset valuations across the sector. Due diligence needs to assess safety governance, not just technology performance.

The Bottom Line

On the Courstide podcast Etan Mark asked: “How do we prevent robots from taking over the world?”

The answer isn’t in science fiction. It’s in the next lawsuit.

When a teenager dies and conversation logs show an AI encouraging emotional dependency, or when a home robot injures a child and sensor logs show the company knew that edge case existed—the question won’t be whether AI is innovative. It will be whether the people building it were responsible.

The attractive nuisance in your living room isn’t just a legal doctrine waiting to be tested. It’s a warning that the old playbook—arbitrate everything, deny duty, blame user choice—won’t work when the “product” walks, talks, learns, and lives alongside vulnerable people.

The companies that survive won’t be the ones with the best arbitration clauses. They’ll be the ones who recognized that building the future means being accountable for it.

Disclaimer: This article is for general information only and is not legal advice.

Concept, drafting, development, meaning making : Katja Maas
Research and editing KM with AI.

This Continuance was inspired by podcast: Courtside with Mark Ferrer Hayden, Ep #6 with Ashley Robinson.

WATCH EPISODE

In the twentieth century, autonomous vehicles lived in our imaginations. In more recent years they lived in the category of almost but not yet. That “almost” phase is ending.

In May 2025, Aurora launched commercial driverless trucking in Texas, operating freight runs between Dallas and Houston with no human driver. This isn’t a pilot, it is commercial service. A major category of transportation is entering an era where the most basic assumption of accident law “someone was driving” may no longer apply.

Autonomous technology is becoming infrastructure. And when infrastructure changes, litigation changes with it.

The Road Is Changing Faster Than the Legal System

The most important misconception about autonomous vehicles is that once the technology improves, the courtroom becomes less relevant. In reality, the opposite is true. When a new system is introduced at scale, the legal system begins testing boundaries of responsibility, probing weak points of accountability, and forcing clarity where marketing materials prefer ambiguity.

This is how society builds standards. Tort law serves as the public’s tool for asking questions press releases won’t answer: Who had control? Who could have prevented harm? Who knew what the risk was?

Autonomous vehicles pose these questions in more complex form because they distribute “control” across people, companies, and decisions made by software rather than human judgment.

It’s Becoming an Industry

Aurora’s announcement marks a threshold, but the shift is industry-wide. Kodiak AI has delivered driverless trucks to customers. Gatik focuses on middle-mile logistics on predictable routes. Waabi is building next-generation systems around simulation.

More companies means more vehicles, more miles, and more edge cases—those rare scenarios that eventually happen to someone.

Robotaxis Bring the Liability Problem to Consumers

Robotaxis may be the first autonomous system to produce mass consumer confusion about responsibility. They collapse three things: the comfort of a private car, the structure of a commercial service, and the fragility of technology.

When someone uses a robotaxi, they’re neither driver nor bystander. They’re a passenger who accepted terms of use. That combination shapes what happens if something goes wrong.

The legal questions become forensic: What did the vehicle detect? What did it classify? What prediction did it make? Litigation becomes less about a single bad moment and more about a chain of system decisions.

The Terms You Sign Matter More Than You Think

Most people treat “terms and conditions” as a formality. But in technology disputes, these agreements determine where a dispute can be brought, whether it can be a class action, and how public the outcome will be.

The most important provision is an arbitration clause. Waymo offers a thirty-day opt-out window. That’s the difference between retaining the ability to bring claims in court versus being required to pursue them in private arbitration. Arbitration provisions often include class-action waivers, preventing collective claims even when the same defect affects thousands.

Most people don’t opt out because they don’t know the choice exists until the window closes.

Brief comic relief from the heavy reading: Eddie Izzard explains why the Terms and Conditions have made liars of us all…

How Arbitration Prevents Public Learning

Arbitration is often framed as efficiency because it’s faster, less expensive. But autonomy isn’t a routine category.

In emerging technology, the legal system isn’t merely assigning blame. It’s generating shared standards for what reasonable safety, oversight, and responsibility look like. That process depends on public rulings and accessible precedent.

Courts create precedent. Arbitration often does not.

Precedent functions like institutional memory, allowing the legal system and the market to build on past conclusions instead of re-litigating foundational issues privately with no durable guidance.

When disputes move out of court, society loses more than a courtroom. It loses a record of what happened and what was learned.

Liability Spreads Across Multiple Parties

In a conventional crash, there’s a central actor: a driver. In autonomous incidents, fault may be distributed across the vehicle manufacturer, autonomy software provider, fleet operator, remote assistance teams, sensor suppliers, mapping systems, and maintenance contractors.

As potentially responsible parties increase, cases become more complex and expensive. That complexity is compounded by a question courts are just beginning to confront: what does “reasonable care” mean when a system makes decisions in milliseconds using machine learning models most consumers can’t interpret and most companies won’t describe?

The New “Black Box” Is Data

Autonomous vehicles generate evidence. Traditional cases depend on human testimony and physical damage. Autonomous cases depend on sensor logs, telemetry, video data, decision sequences, software versions, and system alerts. These sources can clarify facts quickly—or create disputes over access, interpretation, and what counts as proprietary information.

The courtroom won’t disappear. It will become more technical. For lawyers, the ability to understand data trails and litigate autonomy questions will define the difference between an ordinary case and a well-built one.

What This Means

For AV companies: Your arbitration strategy should include proactive consumer education about opt-out windows. The first major public trial will set industry standards—having defensible safety protocols may be a competitive advantage.

For insurers: Autonomous incidents create distributed liability. Traditional single-party coverage is inadequate. The “reasonable care” standard for AI decision-making is undefined—coverage should reflect this uncertainty.

For fleet operators: Document everything. Sensor logs, system updates, and internal safety reviews become the entire case. Companies whose documents show proactive safety governance will fare better.

For investors: The arbitration strategy that protects companies from individual liability may harm the industry by preventing clear safety standards. When courts can’t build precedent, regulators fill the gap more aggressively.

The Bottom Line is In The Air… for now

Driverless trucking is here. Robotaxis are expanding. Autonomous technology is becoming infrastructure. And it won’t stop on the ground. Florida is positioning itself as a testbed for advanced air mobility, aiming to offer commercial air taxi services by late 2026. Archer Aviation has announced plans for a South Florida network connecting Miami, Fort Lauderdale, and Palm Beach. Autonomy is climbing into the air.

Who bears responsibility when control is distributed? How do we prove causation when the decision-maker is software? Can private terms limit public accountability? And most importantly: how does society learn and advance safety standards if dispute outcomes are kept out of court?

References

Aurora Innovation, “Aurora Begins Commercial Driverless Trucking in Texas, Ushering in a New Era of Freight” (May 1, 2025).

Dallas Innovates, “In U.S. First, Aurora Launches Fully Driverless Trucking Deliveries Between Dallas and Houston” (May 1, 2025).

Kodiak AI, “Kodiak AI Delivers Customer-Owned Autonomous RoboTrucks to Atlas… Completes 100 Loads…” (Jan. 24, 2025).

Kodiak AI, “Kodiak AI Delivers More Driverless Trucks to Atlas… Launches Driverless Service Up To 24/7” (June 10, 2025).

Gatik, “We Own the Middle Mile™: Deploying Autonomous Box Trucks…” (Mar. 6, 2023).

NVIDIA Blog, “Waabi Uses Generative AI for Driverless Autonomous Trucking” (July 8, 2024).

Reuters, “Autonomous trucking startup Waabi raises $200 mln” (June 18, 2024).

Reuters, “Waabi ties up with Volvo to roll out self-driving trucks” (Feb. 4, 2025).

Archer Aviation press release (Dec 3, 2025)

Flying Magazine (Dec 3, 2025)

WLRN (Dec 4, 2025)

Courtside Episode 6

Disclaimer:

This article is for general information only and is not legal advice.