In 2026, supervised psilocybin therapy in the US is moving through three state-level models, regulated service centers, time-limited pilot programs and policy shifts that change enforcement or research conditions without building a service system.
Three models showing up across the US
Regulated service centers
A regulated service center model builds a licensed pathway where psilocybin is administered in a controlled setting under a state framework. You will usually see a few repeating parts.
You see a licensed place where services happen. You see licensed roles for the people who support sessions. You often see licensed or registered production and testing steps so the product you administer is tracked, tested and labeled under defined rules. The rules can differ by state, but the overall shape is similar.
Oregon is the clearest example of a state-built service model that is not limited to a medical diagnosis. If you are 21 or older, you can access services through licensed providers under the state framework, with preparation, administration and integration as defined service components.
Colorado is another version of a regulated service model, with its own licensing pathway and service settings under a state framework. It is built around supervised services in designated settings with defined facilitator roles and program rules that cover licensing, conduct and product handling.
When you compare service-center models, you get more clarity by reading the rules as a workflow. Start with who can access services, then where administration is allowed, then how a session is supervised, then how product testing, labeling and tracking are handled. That is the chain that shapes how a program functions in real life.
Time-limited pilot programs
A time-limited pilot program is usually built to answer a narrower question. You typically see a defined time window, a defined setting, a defined population and a defined reporting goal. The point is often program evaluation, feasibility and safety outcomes rather than wide access.
In practice, pilot programs tend to run through established care settings or research-adjacent settings. Eligibility often ties to specific conditions, specific populations, referral pathways or participation criteria written into the law. Oversight and reporting requirements are often more explicit because the pilot is designed to produce a data record that can support next steps.
New Jersey is a recent example of this pilot style, with a statewide pilot approach framed around supervised psilocybin services in controlled settings and an evaluation focus tied to program design and outcomes. If you are tracking 2026 movement, pilots like this signal that lawmakers want structured learning under a bounded scope before broader pathways are considered.
You also see proposed pilots in other states that aim to set up limited access pathways through specific settings, sometimes tied to public health goals, condition lists or defined populations. Bills in this category can change fast across sessions, so you get the best read by focusing on the model features rather than betting on one bill’s final form.
Policy changes without a therapy program
Some states and local jurisdictions change policy without building a supervised therapy program. This can take several forms.
You can see decriminalization or deprioritization policies. You can see reduced penalties. You can see research-oriented policy changes that make it easier to run state-linked research initiatives, create advisory bodies or create grant pathways, even when a supervised therapy program is not created for the public.
If you are trying to understand what these changes mean, focus on what is missing. There is no licensed service setting where administration is authorized as a consumer service. There is no statewide training and licensing framework for facilitators tied to supervised sessions for the public. There may be less enforcement risk in some places, but there is still no public supervised pathway created by the policy change itself.
This model still affects the 2026 picture because it can shape what policy conversations look like next. It can also shape demand signals, advocacy priorities and the direction of future bills. It just does not function as a supervised therapy program on its own.
What is moving in 2026
Newly passed or proposed pilot frameworks
In 2026, you are seeing two kinds of movement at the same time.
First, existing regulated service models keep tightening their operational details. You see updates to training standards, licensure requirements, operating rules and compliance expectations. Oregon published updated program guidance effective January 1, 2026, which is the kind of administrative change that can shift how fast new providers enter and how consistent program delivery becomes.
Second, newly passed or proposed frameworks keep arriving in states that want a narrower pilot or a medicalized access pathway. New Mexico’s medical program is a key example because it is written as a medical framework with approved settings, permitted producers and clinicians, required training and required program data collection. The law also sets a timeline for program implementation and includes a deadline for full implementation by the end of 2027, with public reporting that suggests early patient access goals by the end of 2026.
If you are reading the direction of travel, this is the key point. The regulated service center model is proving that states can stand up a licensing system for supervised administration outside traditional clinical trial structures. The medical and pilot frameworks are showing how lawmakers try to bound eligibility, settings and reporting when they want a more clinical posture.
One constant you still need to keep in view is federal controlled substance scheduling. As of January 2026, psilocybin and psilocin remain listed under Schedule I in federal regulations, which keeps constraints around handling, storage, transport and research compliance in place even as states build their own pathways.
2026 ballot cycle signals
Ballot activity is another source of 2026 signals. A ballot cycle signal does not mean a program will exist in 2026, but it can show you what proposals are gaining traction and how policy framing is changing.
One signal is the move from broad decriminalization language toward language that outlines a regulated access pathway, even if the details are deferred to later rulemaking. Another signal is the use of phased rollouts, where a state may authorize one set of substances or one set of settings first, then broaden later.
Colorado is a clear example of a phased design, with public reporting and policy analysis pointing to expansion elements in 2026 as part of the original framework.
You also see signals from initiative activity in places where proponents are attempting to qualify measures for the 2026 ballot. Alaska is one example where an initiative application related to psychedelics advanced through early administrative steps tied to a potential 2026 vote.
When you interpret ballot signals, treat them as directional. They tell you what may be debated and what models may get copied. They do not tell you what will be operational by a specific date.
How to compare two programs fast
Eligibility and referral
Start with who can access services and how they enter the system. This is the fastest way to see what a program is designed to do.
In Oregon’s service model, adult age is a core eligibility gate, and you do not need a medical diagnosis to access services under the state framework.
In medicalized frameworks, eligibility typically ties to qualifying conditions and a clinician pathway. New Mexico’s law is built around qualified patients, permitted clinicians and approved settings, with program rules designed around medical service standards and patient selection criteria.
In pilots, eligibility can be even narrower. It may be defined by condition, population, setting or program capacity constraints. If you want to compare pilots, you look for the inclusion criteria language and any defined referral rules, then you look for how the program plans to report results.
When you compare two programs, write down the gating steps in one line each. Age gate, diagnosis gate, referral gate, setting gate. That gives you a clear picture without getting stuck in terminology.
Session setting and supervision
Next, focus on where administration happens and what supervision means inside the program.
A regulated service center model usually requires administration in a licensed setting with a trained facilitator present for key parts of the session. Oregon’s framework defines service components and ties administration to licensed service settings and licensed roles, which helps standardize what “supervised” means within that state program.
Colorado’s regulated model similarly centers services in authorized settings, with rules around facilitator roles and operational standards tied to the licensed service environment.
Medical frameworks can define supervision through clinical standards, approved settings and clinician responsibilities. New Mexico’s law points toward treatment protocols, approved settings and medical service standards established through rulemaking, which is a different supervision posture than a non-medical service center model.
If you want to compare two programs quickly, you focus on three details.
You check if administration is limited to a single type of facility or if multiple settings are allowed. You check who is required to be present during administration and what training or licensure they must hold. You check how preparation and integration are defined, because those definitions often control staffing needs and session timing.
Product testing and labeling expectations
Testing and labeling rules shape safety and comparability across programs. They also shape supply realities because they determine what product forms can exist and how product must be verified before use.
Oregon’s model includes rules around product handling and requirements tied to potency, contaminants and labeling expectations within the state framework. Those requirements influence how manufacturers and testing entities operate under the program and they shape what information a client and facilitator see at the point of administration.
Colorado’s rules also set expectations around regulated product handling, labeling and program compliance requirements tied to the natural medicine framework.
Medical frameworks can place similar emphasis on safety protocols, including producing psilocybin from mushrooms, transporting, storing and handling product and administering in an approved setting. New Mexico’s statute explicitly calls for safety protocols and program standards to be established through rules.
A fast comparison method is to ask four questions.
Do you see a required potency test. Do you see required contaminant screening. Do you see a standardized label that includes key content information. Do you see a chain-of-custody or tracking requirement across producer, testing and service setting. Your answers tell you how controlled the product side is, which affects program reliability.
Data reporting and privacy
Data reporting rules tell you what a state wants to learn, and privacy rules tell you how that learning is bounded.
Service models often require some level of reporting for compliance, adverse events and program evaluation. At the same time, privacy protections can limit what gets shared outside the program and how personal information is handled.
Oregon’s statute includes client record and confidentiality provisions tied to its psilocybin services framework, which is a key place to look if you want to understand how privacy is treated in a non-medical service model.
New Mexico’s law explicitly calls for data collection requirements to evaluate the program and to develop best practices, alongside a program structure with permitted roles and approved settings.
A pilot program often leans harder into reporting because the pilot’s purpose is evaluation. In that case, you look for how outcomes are defined, who receives reports, what is public, what stays private and how identifiers are handled.
A quick way to compare is to write down the reporting outputs you can spot.
You note required reports for safety events. You note any required program evaluation report cadence. You note rules for de-identified data, retention periods and who can access records. Then you can compare two programs in minutes.
A one-page comparison checklist
Use this checklist to compare two state models fast. Fill it out in plain words, one line per item.
- Program model
- Service centers, medical framework or time-limited pilot
- Eligibility
- Minimum age
- Diagnosis or qualifying condition required
- Referral required
- Limits by population or setting
- Service setting
- Where administration is allowed
- What types of facilities are allowed
- On-site requirements for safety and monitoring
- Supervision roles
- Who must be present during administration
- Required training or licensure for that role
- Rules for preparation and integration sessions
- Product form
- Allowed product types
- Dose form expectations in program rules
- Testing and labeling
- Potency test required
- Contaminant screening required
- Label content requirements
- Tracking and chain-of-custody rules
- Reporting and privacy
- What must be reported and to whom
- Safety event reporting rules
- Program evaluation reporting cadence
- Privacy and confidentiality rules
- De-identified data rules and retention timing
- 2026 direction
- Rule changes effective in 2026
- Program scale signals
- Ballot or legislative signals tied to 2026
We are Rose Hill Life Sciences, a psychedelic research organization specializing in the production and research of Psilocybe cubensis, operating at the intersection of science and therapeutic integration, and we are based in Massachusetts.
