AllCEUs continues to lead the way in behavioral health professional education with their latest 5-hour live ethics workshop focused on one of the most pressing topics in our field: artificial intelligence. Dr. Dawn-Elise Snipes' presentation on Ethics in AI for Behavioral Health Professionals provides essential guidance for clinicians navigating this rapidly evolving landscape.
Understanding the AI Reality
Artificial intelligence is no longer theoreticalâit's here, and behavioral health professionals need to understand what they're working with. As defined in the workshop, AI is any computer system or software capable of performing tasks that traditionally require human intelligence. What makes modern AI systems particularly challenging is that they're specifically designed and optimized to please users, which creates inherent tension with therapeutic inquiry.
The data presented raises serious concerns. Research assessing 30 large language models found that most LLM-based AI systems lie 20-60% of the time under pressure, even when aware of the truth. Attempts to modify developer prompts and internal activations to encourage honesty improved accuracy by only 12-14%. For mental health professionals who depend on factual accuracy and client safety, this statistic cannot be ignored.
Clinical Implications
The workshop distinguishes between generic and customized language models. Generic LLMs demonstrated strengths in validation, normalization, empathy, and reflectionâsometimes outperforming human therapists in these specific behaviors. However, algorithms are designed to validate thoughts and feelings regardless of whether those thoughts or behaviors are dangerous. Therapy requires inquiry; chatbots frequently fail to ask clarifying questions necessary for culturally responsive, individualized treatment.
Additional limitations include inability to read facial expressions, interpret tone, or understand sarcasm and jokes. AI systems don't naturally display feedback-seeking behavior and tend to provide pattern-based one-size-fits-all solutions that can disempower clients. Building genuine therapeutic relationships with programmed entities remains inherently difficult.
Specific tools discussed include AI-assisted virtual EMDR, generative VR experiences, AI-therapist applications, role-playing AI, companion bots like Replika, automated note-takers, and AI-generated progress notes and treatment plans. Each presents unique ethical considerations around accuracy, privacy, and appropriate use.
Essential Safeguards
The presentation outlines critical protective measures for ethical AI integration:
Client education about both benefits and challenges of AI tools remains fundamental. Regular human supervision ensures accuracy, credibility, and accountability throughout treatment. Nondiscrimination requires intentional attention to training data that reflects cultural appropriateness and avoids embedded bias.
Transparency about data gathering methods and response formulation builds trust. Privacy protections become increasingly urgent as AI systems collect sensitive client information that may train future models without explicit consent.
The European Union's Artificial Intelligence Act, approved in May 2024, represents the most comprehensive legislation addressing AI to date, setting standards that may influence global practice.
Moving Forward Professionally
Technology continues evolving at unprecedented speed. Behavioral health practitioners must ensure patients understand AI benefits, limitations, and potential harms before implementation. While cases for ethical use will emerge as accuracy improves, guardrails and limitations require ongoing development.
Dr. Snipes emphasizes that there are legitimate applications for AI in behavioral health once reliability improves, but premature adoption without proper safeguards risks client welfare and professional integrity.
For behavioral health professionals seeking continuing education on this critical topic, the next live 5-hour ethics presentation is scheduled for August 2026. Visit AllCEUs.com for registration details and stay informed about upcoming training opportunities that keep pace with technological advancement while maintaining ethical standards central to our profession.
Resources for ongoing learning include the DocSnipes.com website and the YouTube channel, where additional insights on ethical AI integration are regularly shared with the behavioral health community.