How Budding Laws On AI And Mental Health Could Go The Route Of ‘Federal Floor, State Ceiling’

AI And Mental Well-Being

As a quick background, I’ve been extensively covering and analyzing a myriad of facets regarding the advent of modern-era AI that produces mental health advice and performs AI-driven therapy. This rising use of AI has principally been spurred by the evolving advances and widespread adoption of generative AI. For an extensive listing of my well over one hundred analyses and postings, see the link here and the link here.

There is little doubt that this is a rapidly developing field and that there are tremendous upsides to be had, but at the same time, regrettably, hidden risks and outright gotchas come into these endeavors, too. I frequently speak up about these pressing matters, including in an appearance on an episode of CBS’s 60 Minutes; see the link here.

AI Providing Mental Health Guidance

Millions upon millions of people are using generative AI as their ongoing advisor on mental health considerations (note that ChatGPT alone has over 900 million weekly active users, a notable proportion of whom dip into mental health aspects; see my analysis at the link here). The top-ranked use of contemporary generative AI and LLMs is to consult with the AI on mental health facets; see my coverage at the link here.

This popular usage makes abundant sense. You can access most of the major generative AI systems for nearly free or at a super low cost, doing so anywhere and at any time. Thus, if you have any mental health qualms that you want to chat about, all you need to do is log in to AI and proceed forthwith on a 24/7 basis.

There are significant worries that AI can readily go off the rails or otherwise dispense unsuitable or even egregiously inappropriate mental health advice. Banner headlines last year accompanied the lawsuit filed against OpenAI for its lack of AI safeguards when it came to providing cognitive advisement.

Today’s generic LLMs, such as ChatGPT, GPT-5, Claude, Gemini, Grok, Copilot, and others (all known as general-purpose AI or GPAI), are not at all akin to the robust capabilities of human therapists. Meanwhile, specialized LLMs are being built to attain similar qualities (known as purpose-built AI or PBAI), but they are still primarily in the development and testing stages. See my coverage at the link here.

Various State Laws On AI Mental Health

A beehive of activity is taking place regarding crafting new AI laws. See my extensive coverage of state-level AI mental health laws at the link here. It is a matter on the minds of the public and in the hands of the state legislators. Some people ardently believe that AI and AI makers are being allowed to run amok. New AI laws are vitally needed to protect society from this onslaught of ubiquitous AI.

I previously examined notable AI and mental health laws passed by Illinois see the link here, one that was also enacted by Nevada see the link here, and one that was enacted by Utah see the link here. Those laws are scoped to prevail within their respective state boundaries. In that sense, these laws are applicable to AI usage within the particular state and do not bear on other states per se.

Big Picture Of AI Mental Health Laws

Not everyone agrees with this pell-mell rush of new AI laws, or at least they are concerned that these AI laws might go overboard. In the zeal to protect society, there is a chance that we might unduly restrict innovation and delay or undercut the benefits of leading-edge AI. The debate is ongoing and heated.

Readers might recall that I proposed a 7-step AI-law-making process that I believe could substantively help regulators to devise new AI laws that are on target and balanced; see my depiction at the link here. This has the added benefit of reducing what I refer to as AI-law legal debt. This refers to AI laws that, though they look shiny, contain hidden debt that must ultimately be paid. Legal glitches and hitches will eventually be found when AI laws are passed without suitable scrutiny and analysis. My prediction is that the slew of newly passed AI laws is likely to create a legal quagmire in the courts.

In terms of the AI laws in the United States, they have not yet stood the test of time, meaning that we won’t really know how well they stand up until there are court cases that test these new laws. It is too early to know whether the laws will survive legal battles waged by AI makers and other contenders. Just because AI laws are enacted does not mean they are proper. All sorts of improper provisions and constitutionally contentious stipulations are undoubtedly buried within these shiny new AI laws.

Congress has repeatedly waded into establishing an overarching federal law that would encompass AI. So far, no dice. The efforts have ultimately faded from view. Thus, at this time, there isn’t an overarching federal law devoted to these controversial AI matters. The big question will be to what degree a sweeping federal law would impact the numerous state-level AI laws. The odds are that many state-level laws would run afoul of a federal mandate, and a tsunami of legal cases would arise as a tussle between federal and state law is undertaken. It surely will be a legal mess.

Accommodating State-By-State Differences

You can likely envision the challenges of the existing and evolving legal landscape governing AI.

Each state does its own thing. The AI law in a state is likely to be poorly specified and be legally ambiguous. States are also amending their AI laws that they previously thought were perfect. Other states that haven’t been enacting AI laws are opting to jump into the waters with both feet. They might borrow wording from other states, change it up, and put it into their legal books.

Going across jurisdictional boundaries when it comes to legal stipulations is a lot harder than other forms of customary localization, such as encompassing currency differences, units of measurement differences, etc. Indeed, in the case of AI, things get extraordinarily tougher. Legal jurisdictional shaping and reshaping of AI go much further. It changes how the AI reasons, what it is willing to say, what questions it asks, what warnings it gives, and, in some cases, what answers it refuses to provide or does provide, all because the governing AI law differs from one state to another.

For my in-depth analysis of the compliance engineering changes required in AI to handle jurisdictional variability, see the link here.

Federal Floor, State Ceiling

Congress is faced with a rapidly growing myriad of state-level AI laws on mental health usage that will be nearly impossible to fit into a yet-to-be-devised comprehensive federal AI law. Conflicts are going to arise. Some areas of AI will be covered by some states, while other states are silent on those matters. Some states are already highly restrictive on some AI aspects, while other states are highly permissive on the same AI aspects. It is a scattergun jigsaw puzzle where the pieces do not fit together.

One belief is that perhaps Congress could focus on a classic regulatory model of a federal floor that is coupled with a state-level ceiling. A comprehensive federal law on AI and mental health would establish a minimum regulatory base. States that happen to already have AI mental health laws below that base would be expected to abide by the stipulated floor. Those states that are already above the floor are presumably good to continue. States with no AI laws in some of those base areas would now have a floor to rely upon.

This is a strategy that acknowledges the existing Byzantine arrangement of state-level AI laws on mental health. If Congress were to instead opt to preempt all those state-level AI laws, doing so would almost certainly generate tremendous hostility from the states, spurring lengthy court action. In theory, a well-shaped federal floor of AI laws on mental health would hopefully end up accommodating most of the states, by and large, plus provide the added flexibility that states could exceed the floor.

That doesn’t mean that the states will necessarily welcome or embrace such an approach. Some states will undoubtedly have grievances about whatever federal floor on AI laws is ultimately enacted. On the federal side, there are indubitably going to be concerns that some states’ AI laws go too far above the floor and exceed any reasonable semblance of a ceiling.

The crux is that either this approach will be a means of reaching across-the-board agreement or it will be rejected as a compromise that isn’t workable.

Regulatory Example Of A Floor

To see how this “federal floor, state ceiling” approach might be devised, let’s look at an example that illustrates the crucial precepts. First, suppose that we identified the existing range of AI laws on mental health and came up with this set of five levels, ranging from least restrictive to most restrictive.

Here are five increasingly restrictive AI laws on mental health chats:

• Level 0: No AI restrictions. No AI law about the use of AI for mental health purposes.
• Level 1: Disclosure of AI. An AI law on the use of AI for mental health requires disclosure to the user that they are interacting with AI.
• Level 2: Human therapist review of AI. An AI law on the use of AI for mental health requires disclosure to the user that they are interacting with AI, plus any mental health assessment by the AI must be reviewed and approved by a licensed therapist.
• Level 3: Human therapist in real-time in AI. An AI law on the use of AI for mental health requires disclosure to the user that they are interacting with AI, plus a licensed therapist must be online and intervening in real-time during any mental health chat.
• Level 4: Ban on AI. An AI law that prohibits the use of AI for mental health purposes.

Level 0 is the least restrictive and consists of no laws about mental health usage in a given U.S. state. The next step up would be Level 1, whereby a U.S. state has a law requiring that the AI disclose to users that they are interacting with AI, doing so whenever the chat veers into mental health guidance. The jump from Level 0 to Level 1 is quite substantial since the AI makers would need to ensure that their AI makes such legally required notifications to the user. The regulation has gone from nothing to something.

But it turns out that merely providing disclosures is still a relatively mild form of restriction. The next step is Level 2. In Level 2, a human therapist must be used to review and potentially approve any mental health recommendations that AI provides to a user. This is quite a further leap into being restrictive. An AI maker would need to have some means of routing AI mental health logs to a licensed therapist for those reviews.

Next, Level 3 is a further level of restriction. A human therapist must be directly in the loop during an AI chat involving mental health guidance. Whereas in Level 2, the therapist did a review, Level 3 requires the therapist to be online and act to intervene in real-time. Finally, Level 4 is an outright ban on the use of AI for mental health. In a given U.S. state that has such a law, the AI can never engage in a mental health chat with users.

Where Is The Floor

Now that we have the levels laid out, the game begins. It is a serious game with significant consequences. Where do you think the most appropriate federal floor should be? You can choose any of the levels.

Is the federal floor to be set at Level 0, Level 1, Level 2, Level 3, or Level 4?

One viewpoint is that Level 0 ought to be the floor. In other words, the federal floor would be that there are no restrictions on the use of AI for mental health. This allows the U.S. states to then decide whether they want any restrictions, and if so, they can choose from Level 1, Level 2, Level 3, or Level 4.

A counterargument is that Level 0 is definitely not a suitable floor. To ensure the minimum safety of the public at large, Level 1 is the least of the several restrictive levels and should undoubtedly become the federal floor. This would mandate that across all U.S. states, the AI would need to disclose to users that they are chatting with AI when discussing mental health guidance.

Whoa, some might say, Level 1 is not nearly enough of a restriction to serve as a floor. Level 2 is really where a true floor ought to be. A human therapist should be legally required to review any AI mental health recommendations to a user. Doing so is the minimally protective action that would be best for the public at large.

I assume you can see how this debate goes. There might be some who would insist that Level 3 is the floor, requiring a therapist to intervene in real-time. And, finally, some would declare that AI should never provide mental health advice; thus, the floor needs to be Level 4.

Conflicts Of The Floor To Existing State AI Laws

An additional complication is that whichever floor is chosen, the odds are that some states will already have existing AI laws that are in direct conflict with the federal law. The thorny question entails what to do about those circumstances.

Suppose a U.S. state had enacted a Level 2 law requiring that a mental health therapist must review any AI recommendation about mental health. Suppose that the federal floor was set at Level 1, consisting of mere AI disclosure. This seems acceptable to all parties since the U.S. state goes above the floor and the federal AI law does not conflict with the U.S. state in that regard. In a sense, the Level 2 in that U.S. state becomes its state-level ceiling and its floor (based on the federal model stating that the state-level restrictions take precedence when exceeding the federal floor).

Switch things around. Suppose the federal floor is Level 2, while the state has already enacted a law mandating Level 1. The federal floor is above the state level. This creates a potential conflict. Presumably, the U.S. state went to a great deal of effort and debate to land on establishing that mere AI disclosure is sufficient. Now, the AI makers providing AI mental health capabilities in that state will need to include a therapist in the loop, despite the U.S. state never requiring that to occur.

Will the U.S. state, in that circumstance, be satisfied to see that the federal floor exceeds its state mandate?

Maybe yes, maybe not. The U.S. state might happily agree with the federal floor and have no trouble with it. On the other hand, the U.S. state might insist that the federal floor is excessive, and the U.S. state should not be hampered by it.

Additional Considerations

I will be covering additional considerations on this topic in an upcoming series.

Some believe that a federal AI law on mental health would need to have a grandfather clause to allow existing U.S. state AI laws on mental health to supersede the federal floor. There are some who believe that the federal AI law should establish national goals about AI and mental health and not legally stipulate anything in a specific form. Just set up broad goals and let the U.S. states each decide what they want to do.

Another big-picture perspective is that the state-level AI laws on mental health should undergo a round of harmonization. This would seemingly be initiated and aided by a federal AI law on mental health. In a harmonization approach, Congress would develop state-level AI law models or templates that the states could readily opt to adopt. States would seemingly examine the templates and ascertain how close or far their existing AI laws extend from those junctures. Over time, the U.S. states would be expected to match their existing AI laws to the templates. Whether this is plausible remains open to doubt, but it is something that is on the table for discussion.

The Path Ahead

A frequent suggestion is that a federal-state AI governance council might be arranged to regularly focus on AI and mental health. This could be established by Congress as a permanent intergovernmental AI council. Members would consist of federal lawmakers, state lawmakers, AI scientists and policy experts, mental health professionals and researchers, and the like.

The federal-state AI governance council might be the first formulation and be used to then devise the federal floor and state-level ceiling framework. Or it could be that the framework is otherwise devised and enacted, and with that comes the council that will work on a go-forward basis henceforth. Each path has its tradeoffs.

A final thought for now. Alan Turing, the famous mathematician and early computer scientist, made this remark: “Codes are a puzzle. A game, just like any other game.” The game afoot right now is: What are we to do about the burgeoning flurry of state-level AI laws on AI usage for mental health, and how will this align with a comprehensive federal AI law on AI and mental health?

It is an enigma that necessitates putting on our thinking caps and finding a viable way to solve the problem and devise a meaningful legal code.