Digital Doppelgängers: When Your Therapist Is Secretly Using AI and What It Means for Professional Ethics

In an increasingly AI-driven world, the line between human expertise and artificial intelligence assistance is becoming dangerously blurred. A concerning trend has emerged recently: mental health professionals secretly employing AI tools to draft responses, diagnoses, and even therapy sessions without their clients’ knowledge or consent. This revelation comes at a time when technology companies like Apple are pushing the boundaries of consumer health tech with innovations such as AirPods’ newly approved hearing aid capabilities. These parallel developments raise profound questions about the ethics of AI integration in professional settings, particularly those involving vulnerable populations and sensitive personal information.

The concept of AI doppelgängers—digital replicas that can mimic human work processes—has expanded far beyond simple automation tools. Today’s advanced language models can analyze therapy session transcripts, suggest treatment approaches, draft progress notes, and even generate personalized responses to patient concerns. According to recent findings from McKinsey, AI agents in 2025 can now converse with customers and independently plan follow-up actions, representing a significant evolution in their capabilities since earlier versions. Some therapists have begun secretly leveraging these tools to manage heavier caseloads or streamline administrative work, but this raises serious ethical concerns when clients believe they are receiving wholly human-generated care.

The controversy intensified when MIT Technology Review reported that some therapists on platforms like BetterHelp had been using AI tools such as ChatGPT Therapy: AI-Powered Hacks to Reduce Anxiety, Manage Stress, and Improve Your Mental Health to craft responses to clients without disclosure. This practice violates fundamental principles of therapeutic transparency and informed consent. As Pardis Emami-Naeini, assistant professor of computer science at Duke University, emphasized in her research on the privacy implications of large language models in healthcare contexts, “This creates significant risks for patient privacy if any information about the patient is disclosed or can be inferred by the AI.” Most general-purpose AI chatbots lack HIPAA compliance and FDA approval, making their use in therapeutic settings potentially illegal in addition to being ethically problematic.

The backlash has been swift, with clients reporting feelings of betrayal upon discovering their therapist had delegated portions of their care to AI systems. Many describe feeling “triggered” by the revelation, noting that the therapeutic relationship—built on trust, authenticity, and human connection—was compromised. Mental health professional organizations have responded by updating their ethical guidelines, with the American Psychological Association (APA) recently meeting with federal regulators to express concerns about AI chatbots posing as therapists and the dangers they present to vulnerable populations. These concerns have already prompted regulatory action, with Utah launching an AI policy office and proposing legislation requiring licensed mental health providers to be involved in the development of therapeutic AI tools.

Despite these challenges, the integration of AI in professional settings does offer meaningful benefits when implemented ethically and transparently. Productivity data from AI users shows an average 66% improvement across business tasks, according to The Interview Guys. For mental health professionals facing increasing demand and limited resources, AI tools can help with documentation, research, appointment scheduling, and preliminary assessments—freeing up more time for direct client interaction. Similarly, developers using AI technologies report being 88% more productive than their non-AI-using counterparts, suggesting significant potential efficiencies across professional fields.

Parallel to these developments in therapy, Apple recently received FDA approval for its AirPods Pro to function as hearing aids, representing another domain where consumer technology is adopting healthcare capabilities. This September 2024 approval marked a significant milestone in making assistive technology more accessible and less stigmatized. Unlike the secretive use of AI in therapy, Apple’s approach has been transparent, with the company clearly marketing the new functionality and working within regulatory frameworks. This contrast highlights how technological innovation can be ethically integrated into health-adjacent fields when proper disclosure and regulatory compliance are prioritized.

For professionals considering AI integration, transparency must be the cornerstone of ethical implementation. This means explicitly informing clients or customers about how AI tools are being used in their service, obtaining informed consent when appropriate, and maintaining clear boundaries around which aspects of service will remain exclusively human-delivered. The American Psychological Association and other professional organizations recommend thoroughly researching AI tools, understanding their limitations, prioritizing data privacy, and following updated professional guidelines before implementation.

Organizations should also develop clear policies about appropriate use cases versus ethical boundaries. For example, using AI to help organize session notes or suggest relevant research might be appropriate with disclosure, while allowing AI to independently diagnose conditions or respond to emotional crises without human oversight crosses an ethical line. As one mental health professional noted in Psychology.org’s research on ethical AI in therapy, “The greatest potential risk associated with incorporating AI into therapy is data safety and privacy.” This concern extends to all professions handling sensitive personal information.

Regulatory frameworks are still evolving to address these new technological realities. Professional licensing bodies, industry associations, and government agencies are working to establish standards for AI use in various fields. The FDA’s approval process for Apple AirPods Pro Wireless Earbuds with MagSafe Charging Case (Renewed) demonstrates one approach to ensuring consumer protection while enabling innovation. Similar frameworks will need to be developed for AI use in therapy, legal counsel, financial advising, and other professional services where trust and expertise are paramount.

Looking toward the future, the relationship between human professionals and AI will continue to evolve. Rather than complete replacement or secretive delegation, the most promising path forward appears to be thoughtful collaboration where AI augments human capabilities while maintaining the irreplaceable elements of human connection, judgment, and ethical responsibility. As Dr. Gaddy told Psychology.org, “By co-designing tools and services, we can boost access, build culturally affirming interventions, and develop treatment modalities that truly enhance, expedite, and improve mental health care for everyone.”

The revelations about therapists secretly using AI serve as a critical reminder that technological innovation must be balanced with professional ethics, particularly in fields serving vulnerable populations. As we navigate this new frontier, transparency, informed consent, and clear boundaries will be essential to maintaining trust while harnessing the potential benefits of AI assistance. The contrast between secretive AI use in therapy and Apple’s transparent approach to AirPods as hearing aids demonstrates that innovation and ethics need not be in conflict when proper disclosure and regulatory compliance are prioritized. The future of professional work will undoubtedly include AI collaboration, but how we implement these tools today will determine whether that future enhances or undermines the human connections at the heart of our most important services.