![[Adobe Stock]](https://sp-ao.shortpixel.ai/client/to_webp,q_glossy,ret_img,w_750,h_375/https://www.psypost.org/wp-content/uploads/2025/03/artificial-intelligence-750x375.jpg)
The integration of artificial intelligence into mental health care has accelerated rapidly, with more than half of psychologists now utilizing these tools to assist with their daily professional duties. While practitioners are increasingly adopting this technology to manage administrative burdens, they remain highly cautious regarding the potential threats it poses to patient privacy and safety, according to the American Psychological Association’s 2025 Practitioner Pulse Survey.
The American Psychological Association represents the largest scientific and professional organization of psychologists in the United States. Its leadership monitors the evolving landscape of mental health practice to understand how professionals navigate changes in technology and patient needs.
In recent years, the field has faced a dual challenge of high demand for services and increasing bureaucratic requirements from insurance providers. These pressures have created an environment where digital tools promise relief from time-consuming paperwork.
However, the introduction of automated systems into sensitive therapeutic environments raises ethical questions regarding confidentiality and the human element of care. To gauge how these tensions are playing out in real-world offices, the association commissioned its annual inquiry into the state of the profession.
The 2025 Practitioner Pulse Survey targeted doctoral-level psychologists who held active licenses to practice in at least one U.S. state. To ensure the results accurately reflected the profession, the research team utilized a probability-based random sampling method. They generated a list of more than 126,000 licensed psychologists using state board data and randomly selected 30,000 individuals to receive invitations.
This approach allowed the researchers to minimize selection bias. Ultimately, 1,742 psychologists completed the survey, providing a snapshot of the workforce. The respondents were primarily female and White, which aligns with historical demographic trends in the field. The majority worked full-time, with private practice being the most common setting.
The survey results revealed a sharp increase in the adoption of artificial intelligence compared to the previous year. In 2024, only 29% of psychologists reported using AI tools. By 2025, that figure had climbed to 56%. The frequency of use also intensified. Nearly three out of 10 psychologists reported using these tools on at least a monthly basis. This represents a substantial shift from 2024, when only about one in 10 reported such frequent usage.
Detailed analysis of the data shows that psychologists are primarily using these tools to handle logistics rather than patient care. Among those who utilized AI, more than half used it to assist with writing emails and other materials. About one-third used it to generate content or summarize clinical notes. These functions address the administrative workload that often detracts from face-to-face time with clients.
Arthur C. Evans Jr., PhD, the CEO of the association, commented on this trend.
“Psychologists are drawn to this field because they’re passionate about improving peoples’ lives, but they can lose hours each day on paperwork and managing the often byzantine requirements of insurance companies,” said Evans. “Leveraging safe and ethical AI tools can increase psychologists’ efficiency, allowing them to reach more people and better serve them.”
Despite the utility of these tools for office management, the survey highlighted deep reservations about their safety. An overwhelming 92% of psychologists cited concerns regarding the use of AI in their field. The most prevalent worry, cited by 67% of respondents, was the potential for data breaches. This is a particularly acute issue in mental health care, where maintaining the confidentiality of patient disclosures is foundational to the therapeutic relationship.
Other concerns focused on the reliability and social impact of the technology. Unanticipated social harms were cited by 64% of respondents. Biases in the input and output of AI models worried 63% of the psychologists surveyed. There is a documented risk that AI models trained on unrepresentative data may perpetuate stereotypes or offer unequal quality of care to marginalized groups.
Additionally, 60% of practitioners expressed concern over inaccurate output or “hallucinations.” This term refers to the tendency of generative AI models to confidently present false or fabricated information as fact. In a clinical setting, such errors could lead to misdiagnosis or inappropriate treatment plans if not caught by a human supervisor.
“Artificial intelligence can help ease some of the pressures that psychologists are facing—for instance, by increasing efficiency and improving access to care—but human oversight remains essential,” said Evans. “Patients need to know they can trust their provider to identify and mitigate risks or biases that arise from using these technologies in their treatment.”
The survey data suggests that psychologists are heeding this need for oversight by keeping AI largely separate from direct clinical tasks. Only 8% of those who used the technology employed it to assist with clinical diagnosis. Furthermore, only 5% utilized chatbot assistance for direct patient interaction. This indicates that while practitioners are willing to delegate paperwork to algorithms, they are hesitant to trust them with the nuances of human psychology.
This hesitation correlates with fears about the future of the profession. The survey found that 38% of psychologists worried that AI might eventually make some of their job duties obsolete. However, the current low rates of clinical adoption suggest that the core functions of therapy remain firmly in human hands for the time being.
The context for this technological shift is a workforce that remains under immense pressure. The survey explored factors beyond technology, painting a picture of a profession straining to meet demand. Nearly half of all psychologists reported that they had no openings for new patients.
Simultaneously, practitioners observed that the mental health crisis has not abated. About 45% of respondents indicated that the severity of their patients’ symptoms is increasing. This rising acuity requires more intensive care and energy from providers, further limiting the number of patients they can effectively treat.
Economic factors also complicate the landscape. The survey revealed that fewer than two-thirds of psychologists accept some form of insurance. Respondents pointed to insufficient reimbursement rates as a primary driver for this decision. They also cited struggles with pre-authorization requirements and audits. These administrative hurdles consume time that could otherwise be spent on treatment.
The association has issued recommendations for psychologists considering the use of AI to ensure ethical practice. They advise obtaining informed consent from patients by clearly communicating how AI tools are used. Practitioners are encouraged to evaluate tools for potential biases that could worsen health disparities.
Compliance with data privacy laws is another priority. The recommendations urge psychologists to understand exactly how patient data is used, stored, or shared by the third-party companies that provide AI services. This due diligence is intended to protect the sanctity of the doctor-patient privilege in a digital age.
The methodology of the 2025 survey differed slightly from previous years to improve accuracy. In prior iterations, the survey screened out ineligible participants. In 2025, the instrument included a section for those who did not meet the criteria, allowing the organization to gather internal data on who was receiving the invites.
The response rate for the survey was 6.6%. While this may appear low to a layperson, it is a typical rate for this type of professional survey and provided a robust sample size for analysis. The demographic breakdown of the sample showed slight shifts toward a younger workforce. The 2025 sample had the highest proportion of early-career practitioners in the history of the survey.
This influx of younger psychologists may influence the adoption rates of new technologies. Early-career professionals are often more accustomed to integrating digital solutions into their workflows. However, the high levels of concern across the board suggest that skepticism of AI is not limited to older generations of practitioners.
The findings from the 2025 Practitioner Pulse Survey illustrate a profession at a crossroads. Psychologists are actively seeking ways to manage an unsustainable workload. AI offers a potential solution to the administrative bottleneck. Yet, the ethical mandates of the profession demand a cautious approach.
The data indicates that while the tools are entering the office, they have not yet entered the therapy room in a meaningful way. Practitioners are balancing the need for efficiency with the imperative to do no harm. As the technology evolves, the field will likely continue to grapple with how to harness the benefits of automation without compromising the human connection that defines psychological care.
