Embracing AI in Therapy and allowing therapists to use AI ethically: New Policies, New Opportunities. Here’s What the Law (and Experts) Say
Introduction: If you’re a therapist in private practice who has started dabbling with AI, you might be wondering about the bigger picture. Is it really okay to bring these new tools into your work? The short answer is yes! Recent moves in U.S. policy – at both federal and state levels – show a pro-innovation stance, encouraging the safe adoption of AI in healthcare. In this post, we’ll explore how new policies (like Executive Order 14179 and OMB memos M-25-21 and M-25-22) are opening doors for AI in therapy, why using AI won’t make you any less ethical or professional, and how AI can actually help you be more effective and energized in your practice. We’ll also tackle common concerns (like energy use and job impact) and give examples of therapist-friendly AI tools (such as heyberries.com/therapists) that make paperwork a breeze. Let’s dive in with a positive, pragmatic look at what this all means for you and your clients. (this blog post is inspired by CCPHC.org Article Here)
AI Policy is Shifting to Encourage Innovation
Not long ago, AI in healthcare was met with caution and piecemeal guidelines. Now, we’re seeing a clear shift: policymakers want to support innovation. A prime example is Executive Order 14179, titled “Removing Barriers to American Leadership in Artificial Intelligence.” Issued in early 2025, this order literally set out to remove roadblocks and boost AI development. It directs the U.S. government to promote AI in ways that advance human well-being and U.S. competitivenessstatic.carahsoft.com. In other words, the federal government is actively urging innovation, not holding it back. The order’s vision is for America to sustain its leadership in AI “to promote human flourishing, economic competitiveness, and national security”: source here – a positive sign that progress in AI is being linked to positive outcomes (like better health and a stronger economy) and allowing therapists to use AI ethically.
Following that Executive Order, the White House Office of Management and Budget (OMB) released two important memoranda – OMB Memo M-25-21 and OMB Memo M-25-22 – in April 2025. These memos provide guidance for federal agencies on using AI. The key thing to know is that they mark a big attitude change at the federal level. According to the White House’s own fact sheet, the new OMB policies shift the government’s approach to a “forward-leaning, pro-innovation, and pro-competition” mindset.source here. In practical terms, Memorandum M-25-21 (on accelerating federal use of AI) and M-25-22 (on efficient AI acquisition) aim to make agencies more agile, cost-effective, and efficient in deploying AI (source here). This is huge: it means federal agencies (including those in health and human services) are being encouraged to adopt AI where it can help. When the policy at the top is pushing for AI adoption, you can bet it will trickle down into everyday tools and practices in healthcare.
The tone from leadership is decidedly pro-innovation. One top official even warned that too much regulation could “kill a transformative industry just as it’s taking off” (source here). This doesn’t mean “no rules at all” (we’ll get to the guardrails soon), but it means the people in charge want AI to thrive. For therapists, that’s good news: the government is basically saying, “We want to foster AI advancements, including in health, so professionals can use these tools to improve services.” In fact, even prior initiatives under the Department of Health and Human Services (HHS) highlighted goals like “catalyzing health AI innovation and adoption” and “promoting trustworthy AI development” (source here). Now, with an even more innovation-friendly stance, those goals are becoming more attainable with strong institutional support.
States Are Joining in with Supportive AI Frameworks
It’s not just the federal government – state policymakers are on board, too. In 2024 alone, state lawmakers introduced nearly 700 AI-related bills across 46 states (source here) many of them touching on healthcare. States like Utah, Colorado, and California have led the way in setting up frameworks to integrate AI into healthcare responsibly (source here). This state-level activism means that wherever you practice, there’s likely movement toward clarifying how AI can be used in care. And importantly, these efforts aren’t about banning AI; they’re about establishing guardrails and guidance that ultimately make it safer and easier for you to use AI tools.
For example, California passed a series of laws in 2024 to protect consumers in AI contexts, including requirements that if a healthcare provider uses AI to communicate clinical information, they must disclose it and offer patients a way to reach a human provider. (source here) That kind of transparency rule is actually helpful – it ensures we use AI openly and ethically, which builds trust with clients. Meanwhile, Utah took a particularly forward-thinking approach: their Artificial Intelligence Act (signed in March 2024) not only put in some basic rules (like telling people when they’re interacting with an AI), but also created an “Artificial Intelligence Learning Laboratory Program” to encourage AI innovation while protecting consumers (source here). In other words, Utah is explicitly saying: let’s innovate with AI in sectors like health, and we’ll figure out how to do it safely. Colorado passed a law focusing on “high-risk AI systems” (which would include things like AI that make healthcare decisions) – setting standards for testing and oversight of those systems (source here). The common theme in these state actions is balance: encourage the beneficial uses of AI in healthcare, but put reasonable safeguards in place so people don’t get hurt or misled.
For therapists, this trend at the state level means you can feel more confident that using AI tools is not some rogue move – it’s actually in line with what regulators envision for the future of healthcare. Policymakers across the spectrum recognize AI can enhance healthcare delivery. The flurry of state laws and bills is basically society’s way of adapting to AI: making sure we reap the benefits (like efficiency and improved access) while minimizing risks. This broad support and guidance translates into a safer environment for you to try AI solutions in your practice. You’re not going out on a limb alone; the system is (slowly) building the safety net and roadmap for integrating AI into care.
Using AI Without Losing Your Ethics or Identity
A big concern among therapists – especially those of us with more liberal or progressive values – is whether embracing AI means betraying our ethical standards or the human-centered nature of our work. Let’s address that head-on: Adopting AI does not mean abandoning your ethics or identity as a therapist. In fact, done right, it should strengthen your ability to uphold those values. The recent pro-AI policies come with an important caveat: do it responsibly. The OMB’s new AI use memo even has “Governance” and “Public Trust” in its title, which shows that the federal guidance explicitly pairs innovation with ethics (source here). The message is: yes, use AI, but use it in a way that maintains transparency, fairness, and trust.
What does this mean for you in practice? It means you can confidently explore AI tools as an extension of your existing ethical framework. Privacy and confidentiality, for example, remain paramount. Many healthcare AI tools are built with privacy in mind (for instance, the AI scribes for therapy notes often tout features like HIPAA-compliance and not storing sensitive data). You’ll still follow all your confidentiality rules – the AI tools just help you do the same tasks more efficiently. It also means you stay in control. There’s a consensus developing that AI in healthcare should augment human professionals, not replace them (source here). As one research article put it, AI is envisioned not as a replacement for human therapists but as a powerful tool that extends the reach of mental health services (source here). In this paradigm, you’re still the decision-maker, and AI is your assistant.
Think of AI as a way to offload the routine, non-glamorous parts of the job, not the core therapeutic relationship. For example, an AI might draft a session summary, but it’s your human empathy and clinical judgment that guide the therapy and interpret that summary. No algorithm can replicate the feeling of connection when a client trusts you with their story, or the creativity you use to tailor an intervention to a unique human being. Those are deeply human skills. AI lacks genuine empathy and moral judgment – qualities only you, as a trained therapist, bring to the table (source here). In recognition of that, experts emphasize AI systems “must be carefully designed and implemented to complement rather than replace the critical human elements of empathy, cultural competence, and nuanced understanding in therapy.” (source here) In short, using an AI tool is not going to turn you into a robot, nor will it turn your therapy into some kind of tech-driven zombie session. You remain a compassionate, ethical practitioner – now with supercharged tools.
It’s also worth noting that many organizations (from professional associations to healthcare institutions) are actively working on ethical guidelines for AI. The goal is to help professionals like you use these tools in line with core principles like beneficence, non-maleficence, autonomy, and justice. So, if you’ve been worried that using AI might mean “selling out” or drifting into morally murky territory, take heart: the trend is to integrate AI in a values-driven way. You can absolutely be a progressive, client-centered, ethically rigorous therapist who also uses AI to enhance their practice. These things are not at odds. In fact, some might argue that part of ethical practice is staying up-to-date with innovations that could help your clients – as long as you vet them and use them wisely.
Common AI Fears
Let’s tackle a couple of the big worries therapists have about AI, and why they’re not as scary as they seem:
- “Will AI replace therapists?” Absolutely not. We’ve touched on this, but it bears repeating with a clear answer. AI is a tool, not a clinician. There’s a saying going around: “AI won’t replace therapists, but therapists who use AI may replace those who don’t.” In other words, the most likely future is one where the best therapists leverage AI to be even more effective. They might get rid of their burnout-inducing paperwork, have data-driven insights at their fingertips, or offer novel digital enhancements to therapy – all thanks to AI – while still providing the empathetic human connection that clients need. Every indication from research and policy is that AI should support and extend the therapist’s work, not substitute for itpmc.ncbi.nlm.nih.gov. Think about it: even the fanciest chatbot cannot truly care for someone the way you do. What it can do is free you from writing the same note for the hundredth time or help you monitor patterns in a client’s progress. This augmented intelligence model is the goal. So, rather than fearing replacement, consider how AI might relieve the aspects of the job that do feel mechanistic or burdensome. You get to spend more time being a therapist and less time being a bureaucrat. Your role might evolve (just as it does with any new tool), but it’s not going away. After all, therapy is fundamentally a human-to-human endeavor, and that’s not changing.
- “Does AI use too much energy / harm the environment?” It’s true that training giant AI models can use a lot of computing power, and there’s been media attention on the carbon footprint of AI. However, let’s put this in perspective. Everyday use of AI (like running a note-taking AI or a chatbot for a few minutes) is not the same as those massive research projects that train a model from scratch. In fact, AI’s energy use right now is just a fraction of the tech sector’s overall energy consumption (the entire tech industry accounts for about 2–3% of global emissions)weforum.org. So, your use of an AI tool is a drop in the bucket of that 2–3%. Moreover, the tech industry is very aware of this issue and is actively working on it. Improvements in hardware and data center efficiency are expected to significantly reduce the energy per AI task over timeweforum.org. Each new generation of chips and servers can do more work with less electricity, and big AI companies are investing heavily in renewable energy to power their data centers. It’s also worth noting the flip side: AI is being used to reduce energy waste in other sectors (for example, optimizing building HVAC systems or improving traffic flow). So there’s a reasonable argument that AI will help offset some of its own footprint by driving efficiency elsewhere. The bottom line for you as a therapist is that using an AI tool for documentation or analysis is comparable to using other common digital services in terms of energy impact. Yes, we should all continue to advocate for green computing and be conscious of sustainability, but you don’t need to feel that adopting AI in your practice is at odds with caring about the planet. You can do both. Use the AI tools, and also support efforts to make them more energy-efficient. (Chances are, the companies behind your AI tools are already on it.)
Those are just two concerns, but the pattern is the same for others: ask questions, get informed, and you’ll often find the fearsome scenario is either unlikely or manageable. Worried about data privacy? Choose AI tools that are built for healthcare and offer strong privacy assurances (and of course, never feed confidential client info into an unauthorized app). Concerned about bias in AI? That’s a real issue being worked on, but if you, as a human professional, are supervising the AI’s contributions, you can catch and correct any odd outputs – essentially, you become the filter that ensures nothing harmful reaches your client. The fact that policies are highlighting “public trust” and fairness means there is recognition at high levels that these issues need addressing, and work is being done to address them. So, staying engaged with AI doesn’t mean blindly trusting it; it means using it critically and constructively, which you’re totally equipped to do.
AI Tools That Make a Therapist’s Life Easier
Enough about policy and theory – what can AI actually do for you, right now, in your practice? The answer is: quite a lot, especially when it comes to the busywork that eats up your time. Over the past couple of years, a number of AI-powered tools have emerged specifically to help therapists and other mental health professionals with their workload. These tools are like having a supercharged assistant who never gets tired of paperwork. Let’s look at one example in depth and then mention a few others:
Meet Berries (http://heyberries.com/therapists) – an AI scribe for mental health professionals. This tool is designed to listen to your therapy sessions (in person or via telehealth) and produce well-organized, thorough session notes for you. Imagine finishing a session and instead of spending 15 minutes typing everything that was said and deciding how to phrase the treatment plan, you get a draft note that’s mostly done, which you can then quickly review and edit. Berries promises exactly that. It’s like having a live scribe in the room, except it’s just software working in the background. Therapists who use it report significant cuts in documentation time – one user even said “No more endless hours to complete notes at the end of the day. I can now focus on what I love, which is providing therapy,” according to a testimonial on their site. The tool is HIPAA-compliant and doesn’t store recordings, meaning it was built with therapy privacy requirements from the ground up. This addresses one of the biggest hesitations therapists have (“Is it safe to discuss sensitive client information?”). With a solution like this, the answer can be yes, because it’s specifically engineered for clinical settings. And Berries isn’t an isolated case; it’s part of a growing trend of companies recognizing that clinical documentation and admin work are huge pain points for providers, and AI is perfectly suited to solve that.
Aside from note-taking, what else can AI do for a therapist? Here are a few examples already out there or on the horizon:
- Progress note generation & treatment planning: Give the AI a few key points or an outline of the session, and it can help flesh out a complete progress note or even suggest a rough treatment plan. You, of course, review and finalize it, but it’s much easier than starting from scratch every time.
- Summarizing client history or assessments: Some therapists are using AI (like GPT-4 via a secure interface) to summarize lengthy intake questionnaires or past records into something more digestible. This can help you quickly grasp what’s going on with a new client or prepare for a session by refreshing key details, without spending an hour sifting through files.
- Administrative assistants: AI scheduling tools can handle appointment bookings or send reminders. There are chatbots that can answer basic FAQs for your practice (like an interactive voicemail/front desk). These might not be therapy-specific, but they lighten the load of running a private practice.
- Therapeutic chatbots and homework: While an AI chatbot isn’t a substitute for therapy, some practitioners set up guided text-based tools that clients can use between sessions for things like mood check-ins, CBT thought challenging exercises, or other psychoeducational interactions. Think of it as an extra resource you provide – the AI can coach someone through a breathing exercise or journal prompt at 2 AM, when you’re not available. Then you can review those logs (if appropriate) and incorporate that into your next session. It’s like having a helper that reinforces the work outside session hours.
- Insights from session recordings: More advanced tools (some still in development) aim to analyze therapy session transcripts (with consent, of course) to highlight themes, track the client’s reported mood over time, or even flag if a client seems to be at increasing risk (e.g., using more language indicating hopelessness). This kind of AI augmented analysis could give you an additional perspective. It’s not that the AI’s judgment is automatically right, but it might show patterns that you can then consider with your own clinical judgment. For instance, if an AI notes “hey, across the last 5 sessions, the client’s tone when talking about work has gotten more negative each time,” that might prompt you to explore that area more next time.
The key with all of these tools is efficiency and support. They’re not doing therapy for you; they’re making the running of therapy and the prep work and follow-up more efficient. By taking off some of that load, you get to focus on what truly matters: the client in front of you and your therapeutic craft. Therapists often wear many hats – healer, administrator, record-keeper, scheduler, billing specialist, etc. AI is there to lighten the administrative and cognitive load of some of those hats, so the “healer” hat can shine brighter.
If you’re new to these tools, a great way to start is to pick one task that drains you the most (say, writing case notes or treatment summaries) and explore an AI tool just for that. You don’t have to overhaul everything at once. Maybe try a free trial of a notes assistant like Berries, or experiment with a secure AI text generator on a de-identified example of a progress note to see how it performs. Many therapists begin by using AI to draft their own documents (like letters or blog posts or curriculum materials) just to get comfortable. As you see the AI’s strengths and limitations, your confidence in where it can help in clinical work will grow. And nothing says you must use AI for anything you’re not comfortable with – it’s there as an option, at your service.
Keep Engaged and Evolve Alongside AI
The landscape of AI in therapy is evolving rapidly – not just the technology itself, but also the policies and best practices around it. Today’s cutting-edge can become tomorrow’s standard tool in surprisingly little time. Staying engaged with these developments is a smart move for any therapist who wants to continue providing the best care possible and keep their practice sustainable. This doesn’t mean you have to become a tech expert or an AI policy wonk, but maintaining a curious and open mindset will serve you well.
Remember, you’re not alone in this journey. Many of your fellow therapists are asking the same questions and taking similar cautious steps into the world of AI. There are communities (online forums, professional association groups, even local meetups or webinars) where therapists share their experiences with AI tools – the good, the bad, and the weird. By participating in those conversations, you can learn tips and tricks, discover what pitfalls to avoid, and also voice your ethical considerations. In fact, it’s crucial that thoughtful therapists (like you) do get involved: your perspective helps ensure that as AI becomes more commonplace, it’s implemented in a way that aligns with our field’s values. If only tech people or policymakers decide how AI should be used in mental health, they might miss something important that a clinician would catch. Your voice matters in shaping the future of AI in therapy.
On the policy front, keep an eye on those federal and state developments we talked about. They will continue to evolve. For instance, agencies might release more specific guidelines for, say, using AI in telehealth or protecting patient data when using third-party AI services. States might refine their laws as we learn what works and what doesn’t. The beautiful thing is that right now the attitude is optimistic and collaborative: policymakers are not looking to punish providers for trying AI, they’re looking to help figure out how to make it work well for everyone. So if you stay informed (through professional newsletters or sites that track health policy), you’ll be able to anticipate changes and stay ahead of the curve. That means less scramble if something becomes a requirement (like, hypothetically, if someday it becomes standard to disclose AI involvement to clients – you’ll already have been doing that because you read it in the California law and adopted it proactively).
Finally, let’s talk about mindset. Approaching AI as a learning opportunity rather than a threat can transform your experience. When you try a new tool, view it as an experiment: “How can this help me and my clients? What do I need to watch out for?” Rather than “This thing better not screw up!” By adopting a bit of a tech explorer mindset, you give yourself permission to play and discover, which is frankly how many of us eventually mastered teletherapy platforms, electronic health records, or even old-fashioned modalities in therapy. AI is just another thing to integrate into your repertoire. And with so many resources and supportive policies backing you up, it’s a friendly environment to give it a go.
In summary, integrating AI into your practice is not about changing who you are as a therapist – it’s about enhancing what you do. You can stay true to your person-centered, socially conscious, ethical approach and take advantage of tools that might make your work easier and your impact greater. The winds of policy are at your back, encouraging this exploration. So, feel reassured and even inspired that you’re stepping into a future where therapists and AI work hand-in-hand to help more people, more efficiently, and maybe with a bit more ease in our day-to-day jobs. This is about empowerment – for you as a practitioner, and ultimately for your clients who benefit from a therapist who isn’t drowning in paperwork or burnt out by administrative fatigue.
Go forth and explore, knowing that the rules are evolving to support you and a whole community of therapists is right there with you, figuring out how to make the most of these new possibilities.
Key Takeaways
Staying engaged with AI is a smart career move. The world of therapy is evolving with technology. By learning and adapting, you’re positioning yourself to serve clients better and keep your practice thriving. You don’t have to be an early adopter of every gadget, but being open to AI – and keeping an eye on evolving guidelines – will help you grow alongside the field. Embracing innovation, with your ethics intact, is a path to both professional fulfillment and improved client outcomes.
Federal and state policies now favor AI innovation in healthcare. Recent moves – like Executive Order 14179 and OMB’s memos M-25-21 and M-25-22 – explicitly encourage adopting AI to improve services, signaling that it’s a good time to innovate in fields like therapy (source here). You’re getting a green light from the top to explore AI tools (with appropriate safeguards in place).
Using AI doesn’t mean compromising ethics or the human touch. On the contrary, the push for AI comes with an emphasis on responsible use. Policies and experts alike stress that AI should augment, not replace professionals, especially in caring fields (source here). You remain in control, and your values of client confidentiality, fairness, and compassion continue to guide how AI is used.
Common fears about AI are manageable. Worried about AI’s carbon footprint? Remember that AI currently accounts for only a small fraction of tech emissions, and efficiency improvements are underway (souce here). Scared of being replaced by a robot? The consensus is that AI is a support tool, not a therapist – it handles routine tasks so you can focus on high-level clinical workpmc.ncbi.nlm.nih.gov. In short, the sky isn’t falling.
AI tools can make your practice more efficient and sustainable. From note-writing assistants (e.g. heyberries.com/therapists for documentation) to scheduling bots, these tools cut down on admin work and free up your time. That means less burnout for you and potentially better care for clients, since you can direct your energy where it matters most.