概述
The therapist then edits each draft to fit a specific client's case formulation. It saves preparation time and makes personalized materials easier to produce, but the therapist still decides what the client sees.
深入探讨
CBT depends on structured materials. Four kinds come up most: Thought record: the client writes down the situation, the automatic thought, the emotion and how strong it was, the evidence for and against the thought, and a more balanced alternative; Exposure hierarchy: a list of feared situations ranked by Subjective Units of Distress (SUDS), usually on a 0-100 scale, so the client can work upward gradually; Behavioral activation: activity schedules and mood ratings that help the client rebuild rewarding routines; and Psychoeducation handouts: plain explanations of ideas like cognitive distortions, the anxiety curve or sleep hygiene. Large language models are good at the parts that are mostly writing. They can rephrase instructions at a chosen reading level, replace generic examples with ones from a client's life, translate, format tables and produce versions for different ages. A useful prompt names the worksheet type, the CBT model it follows (for example, Beck's cognitive model), the client's age and reading level, where the examples should come from, and the output format. AI cannot do the case formulation. It does not know which beliefs keep this client's problem going, which avoidance behaviors matter, or whether exposure is right for the client now. Hierarchies especially need the client's own ratings. A model's guess about which step feels scarier is only a draft for you and the client to reorder. Models also tend to soften exposure steps too much or add safety behaviors, such as bringing a support person, which can undermine the exercise. They may offer generic reassurance where CBT would ask the client to look at evidence. Privacy matters too. Worksheets rarely need identifying details, so describe the client in general terms. Do not paste session notes into a consumer tool. A common misconception is that a polished AI handout is evidence-based because it uses CBT vocabulary. Check it against a trusted source, such as a published treatment manual, before you give it to a client.
战略影响
构建选择
应用级设计决定了人工智能是否能改善实际结果。
团队与工作流程
良好的工作流程集成可以创造用户值得信赖的生产力收益。
风险与安全
范围明确的用例可以减少变更疲劳和实施风险。
The Future of Using AI to Make CBT Worksheets
Worksheet generation will probably move into practice-management and telehealth platforms, where clients can fill in handouts digitally and share them between sessions. That convenience raises new questions about where the data is stored and how much structure clients should get without a therapist present. There is still little research on whether AI-personalized materials improve engagement or outcomes compared with standard worksheets, so treat claims of better results with caution. The skill that will last for clinicians is knowing CBT well enough to notice when a draft drifts away from the model.
现实世界的实施
A teenager with social anxiety loves soccer. The therapist asks AI to rewrite a standard five-column thought record with examples from team practice, at a sixth-grade reading level.
A client with contamination OCD has a list of feared situations. The clinician gives AI the list without identifying details and asks it to order them into an exposure hierarchy with blank SUDS (0-100) columns. Then the clinician and client reorder the steps together in session.
A therapist generates a one-page handout in Spanish that explains the cognitive triangle of thoughts, feelings and behaviors. A fluent colleague checks the translation before anyone uses it.
A depressed client works nights. AI drafts a behavioral activation schedule that starts at a 3 p.m. wake time instead of following the usual 9-to-5 template.
风险与防护栏
将损坏的流程自动化可能会加剧现有问题。
团队可能会过度自动化并消除所需的人工判断。
如果不持续评估输出，质量可能会出现偏差。
实施路线图
绘制当前工作流程并确定摩擦最大的步骤。
在完全自动化之前定义人工检查点。
对用户进行提示、升级路径和质量标准方面的培训。
跟踪任务级结果以确认持续价值。
不断探索
Free newsletter
Get the daily AI briefing
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Take the Using AI to Make CBT Worksheets quiz
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
常见问题
What is Using AI to Make CBT Worksheets?
Using AI to make CBT worksheets means asking a language model to draft thought records, exposure hierarchies, behavioral activation plans and psychoeducation handouts. The therapist then edits each draft to fit a specific client's case formulation. It saves preparation time and makes personalized materials easier to produce, but the therapist still decides what the client sees.
On an exposure hierarchy, what does a SUDS rating measure?
SUDS stands for Subjective Units of Distress. Clients rate how distressing each situation feels, usually from 0 to 100, and the ratings set the order of the hierarchy.
Which part of making CBT worksheets does the guide say AI cannot do?
AI can handle wording and formatting. Understanding what maintains a particular client's problem takes clinical assessment that the model does not have.
Why does the guide warn about AI-drafted exposure steps?
Safety behaviors and steps that are too gentle can undermine exposure. That is why the guide says to remove them when you review the draft.
When building a hierarchy, whose ratings should decide the final order of steps?
The guide says hierarchies need the client's own ratings. The model's ordering is only a draft for the therapist and client to reorder together.
A therapist wants a thought record for a 14-year-old soccer player. Which prompt element does the guide recommend?
The guide recommends naming the audience's age, reading level and where examples should come from, and it warns against pasting session notes into consumer tools.
继续学习
相关指南
为此主题精选的更多指南