Columns

Financial reforms adopted in the wake of the 2007–2008 global financial crisis have not prevented continuing bailouts, Professor Jeffery Zhang points out in a new paper. 

By Bob Needham
Michigan Law

Financial reforms adopted in the wake of the 2007–2008 global financial crisis have not prevented continuing bailouts, Professor Jeffery Zhang points out in a new paper. 

Zhang refers to the problem as “the Mike Tyson theory of financial regulation,” after the boxer’s famous saying, “Everyone has a plan until they get punched in the mouth.” In this case, the Dodd-Frank reforms that looked good on paper have proven to be “too narrow in design and too daunting in practice,” Zhang writes.

His new paper, forthcoming in the Iowa Law Review, explores the reasons that Dodd-Frank has not performed as hoped and considers an alternative. He recently answered five questions on the issue:

1. How common are bank failures these days?

Individual bank failures happen all the time. As I try to highlight in the paper, this is nothing new. In fact, we have a well-oiled system for dealing with these failures. 

Supervisors are good at coming in and saying, “Hey, time to close up shop.” Then the FDIC comes in and makes sure all the deposits are good. There are no hiccups for individuals and minimal hiccups for the economy.

However, the worry coming out of the global financial crisis was that the system was on fire, not just one individual bank. Moreover, it was these really large banks and other financial institutions that were wobbly. So the idea was that maybe we should do more to prevent that from happening without resorting to bailouts.

2. You discuss two particular provisions in Dodd-Frank designed to prevent taxpayer bailouts. What are they?

Dodd-Frank in general was a response to the ad-hoc bailouts in 2007 and 2008. The whole idea was to improve the safety and soundness of the entire system and of these individual institutions so that we don’t have these ad-hoc bailouts in the future. 

Title I of Dodd-Frank requires the largest institutions to write “living wills”—a plan to resolve a bad situation without asking for government intervention. Title II is a complementary measure that says if Title I proves insufficient, then the FDIC can step in.

In practice, though, these remedies have never been used. It was a worthwhile experiment. I’m all in favor of regulatory experimentation, but if we’ve tried something, after a while we should honestly reassess the weak points and the strong points.

3. What went wrong?

A fear of the unknown. We established these tools, but we’ve never used them. If a house is on fire, and you have no experience with putting out fires, your knee-jerk reaction is to run away. 

You’re not going to use a new tool during a panic. 

That’s where Mike Tyson comes in: Everyone has an elaborate plan for a crisis—but if you don’t practice, you get a little scared. You abandon the plan and fall back on the familiar. In financial crises, that means resorting to more bailouts. 

We saw this in 2023 with the bailouts of Silicon Valley Bank and Credit Suisse. Regulators spent 12 and a half years creating a regulatory framework to not make the same mistakes again. Then you get the stress test you’re looking for in both the US and in Switzerland, and the plan got thrown out the window. 

If you ever want this resolution framework to be credible in the eyes of the market, to use it successfully, you’ve got to put it into practice somehow. Otherwise, every time a crisis comes up, we’re going to get that knee-jerk reaction.

4. Your paper suggests that part of the solution is to “practice” on smaller institutions?

Yes—institutions that are still large but not massive. If regulators could use the tools on smaller institutions, they could build confidence but also demonstrate that the tools can work. 

Right now, this won’t happen because the thresholds for invoking the tools are so high. We have to lower the thresholds—I’m suggesting institutions of somewhere between $10 billion and $50 billion. Those are big enough to matter, but not “too big to fail.” 

5. So if the threshold was lowered, would that be enough, or do we need further changes?

If we really want to make the system safer, that by itself is not enough. If we can lower the threshold, that at least gets us in the right direction. 

One other change that might help is if we direct the regulators to actually use the tools. 

Congress might have to revise the regulation to make it an automatic process—requiring that unless otherwise determined, you must use the resolution regime. Changing that structure could 
help.

The word “niche” carries several meanings, from describing a small recess in a wall to referring to a specialized segment of the market. The most powerful meaning, however, is the one that speaks to a sense of purpose, being “best fitted” for something. In this sense, a niche is not just a comer carved out but a place where a person’s skills, passions, and values align in a way that feels natural and sustaining.

By Sarah E. Kuchon
Hohauser Kuchon

The word “niche” carries several meanings, from describing a small recess in a wall to referring to a specialized segment of the market. The most powerful meaning, however, is the one that speaks to a sense of purpose, being “best fitted” for something. In this sense, a niche is not just a comer carved out but a place where a person’s skills, passions, and values align in a way that feels natural and sustaining. We often speak of”finding our niche” as if each of us has a single destination waiting to be discovered. However, a life of meaning is not about narrowing toward one fixed point. It is about finding a space that allows us to grow, to give, and to remain true to who we are. Whatever direction we choose, the goal is the same: to bring the best version of ourselves forward and to be “best fitted” for what we take on. 

What It Means to Be “Best Fitted”

Being “best fitted” goes beyond ability. Skills can be learned and refined, but a true niche is about alignment between one’s inner self and outer work. Being “best fitted” is not about landing a perfect job. It is about identifying where what we do is both meaningful to us and valuable to others. It is about making intentional choices about how we spend our time and energy. It is also about aligning our work, commitments, and relationships with our strengths, passions, and values rather than drifting into roles or paths by default. When we are best fitted, we thrive. 

The Cost of Being Misfitted 

Too often, professionals stumble into careers without pausing to ask whether they have found their real fit. They may achieve outward success yet still feel restless or unfulfilled. The cost of being misfitted is subtle but profound. When our work does not align with who we are, it often shows up as restlessness, fatigue, or a nagging sense that something is missing. Success achieved without alignment can feel hollow, leaving us wondering why the milestone we chased did not bring the fulfillment we imagined. Over time, the strain of forcing ourselves into roles that do not fit can erode not only our energy but also our confidence, creativity, and joy.

Discovering Where You Are Best Fitted 

Finding a role, a path, or even a way of living that aligns with our strengths, passions, and values is rarely accidental. It is not a matter of luck or waiting for the perfect opportunity but a process of discovery that unfolds through curiosity, reflection, and experimentation. Often, small, intentional steps are needed to illuminate the path. The following strategies can provide a framework for discovery, turning what might feel like aimless wandering into thoughtful exploration. 

—Reflect on Your Strengths and Passions 

Begin by taking stock of the activities and roles that energize you. Ask yourself: “What do I do well? What comes naturally to me?” These are your strengths. Then ask: “What do I enjoy doing so much that time seems to fly by when doing it? What projects leave me feeling creative and engaged?” These are your passions. The place where your strengths and passions overlap is often the first clue toward a niche that feels authentic and meaningful. 

—Notice Patterns in Feedback 

Next, pay attention to how others respond to your contributions. Colleagues, friends, and clients often notice our gifts before we do. 

Ask yourself: “Which of my skills or qualities are consistently recognized or appreciated?” External perspectives and feedback can shine a light on strengths we might overlook, helping us see where our natural abilities meet the needs of the world around us. 

—Experiment and Explore 

Finding our niche is rarely a one-step decision or a single choice. It is something we discover through trial, reflection, and adjustment. Be willing to step outside your comfort zone and try new roles, projects, or experiences. Each experience gives you data. Ask yourself: “What energizes me? What drains me? What feels aligned? What feels mismatched?” Each experiment is a chance to test and explore what feels aligned and what does not. 

—Align with Values and Market Needs 

A true niche honors our values and meets real needs. You may have passion and skill in an area, but if there is no demand, it will be difficult to build a career around it. On the other hand, working in an area of high demand that does not align with your values will leave you burned out and unfulfilled. The sweet spot lies where authenticity meets opportunity, when your work reflects who you are and contributes something meaningful. To identify your values, start by considering what matters most to you. If the gap between your values and work is too wide, it might be time to seek a new role, organization, or field that offers a better fit. This can feel risky, but long-term fulfillment often requires the courage to step away, explore, and adjust. 

Finding My Niche 

For years, I felt like I was just moving through life. As a personal injury attorney, I had honed my abilities and achieved success, yet something felt incomplete. I wanted to serve my clients in a deeper, more holistic way, one that went beyond the confines of law. In pursuing that purpose, I explored counseling and yoga, developing new skills and discovering unexpected ways to connect with and serve my clients. 

While some thought I was taking a risky or unconventional path, to me it felt like the only one. In following my passions, I began to understand what it truly meant to be “best fitted.” At first glance, being an attorney, a counselor, and a yoga instructor seem like three very different paths, but for me they share deeper, meaningful connections. In essence, all three roles center on service, guidance, and human connection, using specialized knowledge to help people navigate challenges, grow, and find alignment in their lives. What once felt like separate callings have woven together into a life of purpose, where I discovered my niche of serving others with both skill and heart. 

Be Patient with the Process 

The search for a niche is really the search for ourselves. It is less about finding a fixed destination and more about cultivating awareness of where we are best suited to contribute, grow, and thrive. Whether we wander and explore or hone a particular path, the journey itself teaches us who we are and what matters most. In this way, the pursuit of a niche becomes not just a search for alignment but a practice of presence, reflection, and authenticity. 

Through small reflections, experiments, and observations, we can gain a clearer understanding of where we are “best fitted,” thereby finding our niche. 
————————
Sarah E. Kuchon, of Hohauser Kuchon, is the 93rd president of the Oakland County Bar Association.

In today’s digital world, artificial intelligence (AI) tools are everywhere. Many homeowners, business owners, and policyholders turn to AI chatbots for quick answers about insurance coverage, insurance claims, or legal questions about insurance law. While AI can provide helpful general information in limited circumstances, it is not a substitute for legal advice from an experienced insurance attorney. 

By Rabih Hamawi

In today’s digital world, artificial intelligence (AI) tools are everywhere. Many homeowners, business owners, and policyholders turn to AI chatbots for quick answers about insurance coverage, insurance claims, or legal questions about insurance law. While AI can provide helpful general information in limited circumstances, it is not a substitute for legal advice from an experienced insurance attorney. 

Mistaking AI guidance for professional counsel can lead to costly mistakes when filing insurance claims or dealing with insurers.

Why AI Cannot Give Legal Advice

AI tools, including ChatGPT, Bing AI, Google Gemini, Microsoft Copilot, Claude (Anthropic), Perplexity AI, and Grok (xAI), can generate responses based on patterns in data, but they cannot:

• Analyze your personal or commercial insurance policy in detail

• Interpret complex legal language specific to your jurisdiction

• Provide advice tailored to your unique situation

• Represent you in disputes with insurance companies

• Represent you in court when suing an insurance company

Only an experienced insurance attorney can review your policy, evaluate coverage, and guide you on your legal options. Using AI as a sole source of guidance can create misunderstandings or lead to denied claims.

Common AI Misunderstandings About Insurance Coverage 

Many policyholders assume AI can fully explain insurance coverage, but it often provides only general information. AI cannot review your specific policy, identify exclusions, or confirm whether a particular loss is covered.

Relying solely on AI can lead to mistakes, such as misunderstanding coverage or missing critical claim requirements. This often causes confusion, such as:
Coverage assumptions – AI may generalize insurance terms, but cannot confirm if your policy includes specific protections.

Policy exclusions – AI might not detect subtle exclusions in your policy, which insurers can use to limit or deny claims.

Claim procedures – Filing a claim incorrectly can reduce your reimbursement or void coverage. AI can suggest steps, but these may not match your insurer’s requirements.  

Example: A homeowner asks AI if flood damage is covered. AI responds based on general knowledge: “Some homeowners insurance includes flood coverage.” In reality, most standard policies exclude flood damage, and only a flood policy or endorsement provides coverage.

How Insurance Coverage Appears in Your Policy

Insurance coverage is typically structured as:

Declarations Page – Summarizes your coverages, limits, and deductibles

Insuring Agreement – Explains what your insurer promises to cover

Exclusions – Lists perils not covered

Conditions – Details requirements to maintain coverage (e.g., notice deadlines, documentations, submitting to an examination under oath, etc.)

AI can explain these sections generally, but its explanation should not be construed as legal advice. AI cannot tell you whether your specific incident is covered or whether you have complied with the policy conditions. 

How Insurers Often Limit or Deny Coverage

Insurance companies often deny or reduce claims based on:

Policy exclusions – Damage or events not covered in the policy

Late notice or improper filing – Missing deadlines or submitting incomplete forms

Documentation gaps – Lack of evidence proving loss or damage

Policy misinterpretation – Insurers may argue your claim falls outside coverage 

Relying solely on AI increases the risk of common AI misunderstandings about insurance coverage, which can lead to denied claims.
 
Practical Mistakes to Avoid When Using AI for Legal Questions

Here are the most common mistakes policyholders make when relying on AI:

Assuming AI is a licensed attorney – AI cannot represent you or provide legal advice.

Skipping policy review – Blindly following AI guidance without reading your policy can lead to coverage mistakes and claim denials.

Ignoring deadlines – AI might not highlight critical timelines in your claim process.

Overlooking exclusions – AI cannot detect nuanced exclusions that insurers exploit.

Sharing sensitive information – Inputting policy numbers or personal details into AI tools may compromise privacy.

Do not use AI except for basic explanations, and always verify coverage details with an experienced insurance attorney.

Attorney & Counselor Rabih Hamawi has extensive expertise in insurance coverage, business negotiations, and commercial litigation. He focuses his practice on representing businessowners, homeowners, property owners, and other insurance policyholders in fire, property damage, and insurance-coverage disputes with insurance companies and in errors-and-omissions cases against insurance agents. Reach Law Office of Rabih Hamawi at (248) 905-1133.

Many policyholders who have disputes with their insurers, and decide to sue them, often find themselves asking the same pressing question: “How long will this lawsuit take and what can I expect?” It’s a natural concern, as insurance claims—especially those involving significant property damage, fire losses, or denied insurance coverage—can have a major impact on an insured-plaintiff's finances, daily life, and peace of mind.

By Rabih Hamawi
Law Office of Rabih Hamawi P.C.

Many policyholders who have disputes with their insurers, and decide to sue them, often find themselves asking the same pressing question: “How long will this lawsuit take and what can I expect?” It’s a natural concern, as insurance claims—especially those involving significant property damage, fire losses, or denied insurance coverage—can have a major impact on an insured-plaintiff's finances, daily life, and peace of mind.

While every lawsuit is unique and timelines can vary depending on the complexity of the case, insurance lawsuits typically take at least one to two years, not including appeals. 

This is due to the multiple stages involved in litigation, including filing the complaint, exchanging evidence, engaging in discovery, participating in mediation or other alternative dispute resolution, and potentially going to trial. By understanding these stages and what to expect at each step, plaintiffs can set realistic expectations, plan accordingly, and remain proactive throughout the legal process.

Preparing for a Successful Insurance Lawsuit

Success in an insurance lawsuit doesn’t start in the courtroom—it starts long before you file. It starts when you report a claim for the very first time. Preparation is everything. This means that as a start, you must carefully review your policy, organize all correspondence with your insurer, gather photos and videos, repair estimates, invoices, and any expert reports that support your claim.

Understanding the strengths and weaknesses of your case allows you to anticipate challenges and respond effectively. Being thorough at this stage not only strengthens your position but also sets the tone for the entire litigation process, giving you confidence and control as you move forward. The more prepared you are, the more likely your case will proceed smoothly and increase your chances of a favorable outcome.

Filing the Lawsuit 

The process begins when your attorney formally files a complaint with the court. In an insurance lawsuit, this usually involves claims for property damage, fire loss, or denied insurance coverage. Once filed, the insurance company is then officially notified and served, and it is required to respond, by answering the complaint.

The Answer and Preliminary Motions

After serving your complaint on you insurer, it typically has a set period of time to file an answer. When it answers, the insurer admits or denies your claims and may raise defenses. At this stage, either party may also file preliminary motions, such as motions to dismiss, which can slightly extend the timeline.

Discovery Phase

The discovery phase is one of the most time-intensive parts of litigation. During discovery, both sides exchange evidence, documents, and witness information. Depositions, interrogatories, and requests for production help build each party’s case. In insurance disputes, this phase can take several months or even over a year, especially if experts are involved, such as engineers or fire investigators.

Pre-Trial Mediation and Motions

Even before a trial, there are often opportunities to resolve the case. Settlement negotiations or mediation can sometimes resolve disputes faster. But if negotiations fail, parties may file pre-trial motions to clarify issues, exclude evidence, or request summary disposition or judgment. Each motion can add weeks or months to the process.

Trial

If the case proceeds to trial, the court schedules hearings and trial dates, which can be influenced by the court’s docket. A typical trial may last several days to weeks, depending on the complexity of the case.

Post-Trial and Appeals

After the trial, either party may file appeals, which can extend the resolution timeline by additional months or even years. But even without appeals, most insurance disputes take at least two years from filing to resolution, and sometimes more.

Key Takeaways 

Insurance lawsuits are often complex and involve detailed evidence and expert testimony.

The process typically lasts one to two years, or more.

Understanding each stage—filing, discovery, pre-trial motions, trial, and possible appeals—helps policyholders stay prepared and avoid surprises.

If you are dealing with a denied insurance claim or ongoing insurance dispute, working with an experienced insurance attorney can streamline the process, ensure your rights are protected, and help you pursue the compensation you deserve.
 –––
Attorney & Counselor Rabih Hamawi has extensive expertise in insurance coverage, business negotiations, and commercial litigation. He focuses his practice on representing businessowners, homeowners, property owners, and other insurance policyholders in fire, property damage, and insurance-coverage disputes with insurance companies and in errors-and-omissions cases against insurance agents. Law Office of Rabih Hamawi can be reached at (248) 905-1133. 

Very few business lawsuits arrive without warning. In practice, most disputes give off clear signals long before a demand letter appears or a complaint is filed. The difficulty is not that those signs are hidden, but that they are easy to rationalize away when you are focused on running a company.

By Zana Tomich
Dalton &?Tomich PLC

Very few business lawsuits arrive without warning. In practice, most disputes give off clear signals long before a demand letter appears or a complaint is filed. The difficulty is not that those signs are hidden, but that they are easy to rationalize away when you are focused on running a company.

After years of working with businesses as outside general counsel, a pattern emerges. The same behaviors, the same shifts in tone, and the same breakdowns in relationships tend to precede litigation. When those signals are recognized early, many disputes can be resolved quietly. When they are ignored, the path to court is often difficult to avoid.

One of the earliest indicators is a sudden change in communication. When a customer, vendor, partner, or employee who was once responsive begins to go quiet, something is usually happening behind the scenes. Emails go unanswered. Calls are returned late, if at all. Meetings are postponed or canceled. The tone, once informal and collaborative, becomes careful or distant.

People rarely disengage when they feel satisfied or secure. Silence is often strategic. It allows time to gather information, review documents, and seek advice without alerting the other side. In many disputes, this quiet period marks the transition from frustration to preparation.

Another common warning sign is a noticeable decline in contract performance. Missed deadlines, inconsistent quality, or unexpected disputes over invoices often signal that a business relationship is under strain. In some cases, one party begins reinterpreting the scope of work or insisting on contract terms that had previously been applied loosely or not at all.

As performance drifts, trust erodes. Once trust begins to break down, parties tend to document more aggressively. Emails become longer and more formal. Minor issues are memorialized. Phrases like “that’s not what we agreed to” appear with increasing frequency. This shift toward documentation is rarely accidental; it often reflects a growing concern that the relationship may not end cooperatively.

Employment disputes follow a similar pattern, though they usually begin on a more personal level. Employees rarely frame concerns in legal terms at the outset. Instead, they express feelings of unfairness or being singled out. Statements such as “I don’t feel supported,” “this feels like retaliation,” or “management doesn’t treat people equally” often appear before any formal complaint is made.

Even when management disagrees with the employee’s perspective, these statements matter. They signal that the employee is beginning to view workplace issues through the lens of rights and protections. 

Without careful handling, what starts as a workplace grievance can evolve into a claim that carries legal and reputational consequences.

Another strong indicator of impending conflict is a sudden insistence on documentation. Customers who previously accepted work without question may begin demanding detailed reports, timestamped records, or strict adherence to contractual procedures. Requests for clarification about conversations that occurred months earlier may surface unexpectedly.

This change is rarely about organization alone. More often, it reflects an effort to build a record; either to justify withholding payment or to support a future claim. When this happens, it becomes especially important for a business to ensure that its own records are accurate, complete, and consistent.

Breakdowns with vendors or partners also tend to show themselves early. A party who begins ignoring payment terms, confidentiality obligations, or performance standards may be experiencing financial strain or reevaluating the relationship. In other cases, they may believe that the other side breached first. Once formed, that belief often becomes the foundation of a legal dispute.

One of the more subtle signals appears in casual conversation. When someone mentions having spoken to a family member or friend who is a lawyer, or raises concerns about whether something is “legal,” the dynamic has shifted. These comments suggest that legal options are being explored, even if no formal steps have been taken.

Finally, there is the instinctive sense that something is off. Business owners are often quick to dismiss that feeling, telling themselves that tensions will pass or that long-standing relationships will prevent escalation. 

In hindsight, many disputes can be traced back to a moment when a concern was noticed and then set aside.

When these warning signs appear, timing matters. Issues addressed early can often be resolved with a clarifying conversation, a written adjustment, or a modest course correction. Left unattended, the same issues tend to harden into positions that are difficult to unwind.

Litigation is rarely the product of a single event. More often, it is the end result of a series of missed opportunities to intervene. Businesses that pay attention to the early signals are better positioned to protect their operations, relationships, and resources before conflict becomes unavoidable.

Medical applications of artificial intelligence usually require a clinician to be involved—but clinicians may be ill equipped to fill oversight roles, Professor Nicholson Price argues in a new paper.

By Bob Needham
Michigan Law

Medical applications of artificial intelligence usually require a clinician to be involved—but clinicians may be ill equipped to fill oversight roles, Professor Nicholson Price argues in a new paper.

Clinicians are often unavailable, ineffective, or both, Price writes in the Emory Law Journal. Alternative structures will be necessary for AI to assist in providing quality patient care, he says.

Price recently answered five questions on the issue:

1. Your article focuses on the practice of using a person to ensure that AI is working properly. Why is this important?

This paper draws from a piece that I wrote with Rebecca Crootoff at the University of Richmond Law School and Margot Kaminski at the University of Colorado Law School on the concept of “Humans in the Loop.” What this means is that in a specific decision, there is a human who is involved in that decision. It’s not just left up to the AI. 

For example, in a medical setting, if you had a patient who took an image of a lesion on their arm and an AI system analyzed that and came back with a diagnosis that said the lesion is precancerous, that decision would not have a human in the loop. If instead there’s an image of the lesion and an AI system flags it as potentially precancerous and a dermatologist confirms that it’s precancerous, we’ve added a human into the decisional loop.

This is important because it’s a real go-to fix for situations where an algorithm might not get everything quite right. Maybe the algorithm is not right for a particular patient. Maybe it’s biased, maybe it has errors, maybe it’s not kind. Whatever problem you have within an algorithmic system, a frequent solution is to put a human in the loop to make sure the algorithm is performing correctly. 

2. You note that one problem with doing this in medical settings is that clinicians are not particularly effective in this role. Why not?

Empirically speaking, clinicians—at least as far as we’ve been able to tell—often aren’t great at catching AI errors. Double-checking requires time, and time is something a lot of clinicians just don’t have. 

Another part of this is automation bias, which arises when you’ve got a system in which a machine is giving you answers over time. If it does a pretty good job, you’ll tend to defer to the machine. 

Then there’s a question of training. We are training young doctors to look at these systems and identify algorithmic errors. There’s a program at Michigan Medicine (DATA-MD) that helps clinicians understand how to use medical AI, what to look for, what problems to be aware of. But there are lots of physicians who were trained a while ago, and this was just not part of their training.

3. The other issue you discuss is that clinicians are not necessarily even available. How does that play out?

It’s easy to look at the question of a clinician in a loop and think, how is this AI system going to interact with a trained expert doctor who has the time to interact with it? That’s certainly an important question to ask. Yet frequently, we’re just not going to have that person around. 

One of the potential benefits of AI is the ability to extend care to folks who otherwise don’t have access to it. But if you expect that that care is going to be supervised by a specialist, you’re just recreating the initial problem. If you have a tool that can do something you might have needed an ­ophthalmologist to do—but the tool only works if you have an ophthalmologist to double-check the answers—you just lost the value of the tool. 

There are ways to try to deal with some of these issues, but the idea of AI ­driving broad access to care is really ­challenging to square with the requirement that we have an expert ­double-checking these answers.

4. What should be done differently?

First, just be aware that this is a problem. Slapping a human into the loop is a deeply ingrained and easy solution, but it’s not a good solution in many situations. 

Second, to the extent that we do have humans to be in the loop—which I recognize they will be, for a long time, in lots of contexts—you need to enable them to succeed. That means identifying what you expect the humans to do. If you say, “I want you to explain the results of this to patients and be kind and empathetic,” 
that’s one important thing a human can do. If you say, “I want you to double-check these results and make sure they’re accurate,” that’s a different thing.

Third, we need monitoring to make sure the system—the combination of human and AI—is performing well over time. 

5. You also suggest standards for systems and governance?

Yes. One of the challenges that arises in this space is designing systems. The places that’ll be able to spend the most effort on getting that right are places like Michigan or Harvard or Memorial Sloan Kettering—places that have lots of resources. 

But what if we really want, in addition to those places getting better, to enable lots of care for lots of folks? There’s a real role for nongovernmental organizations, policy makers, and academic medical centers to work together to develop standards and best practices. 

There is some effort to develop those frameworks happening now. It’s really important that that work continue and be supported and be widely distributed to help enable all sorts of different places to have AI and humans working well together—or to just recognize that sometimes you’re not going to have a human in the loop at all.