Designing with Love

AI Tutors, Coaches, and Practice Bots: When They Help and When They Don't

13 min · I går
episode AI Tutors, Coaches, and Practice Bots: When They Help and When They Don't cover

Beskrivelse

A bot can sound warm, responsive, and confident, and still teach the wrong thing. That’s the tension we dig into as we explore AI tutors, AI coaches, and practice bots through the lens that matters most to instructional designers: practice design that improves real performance, not just chat transcripts. Jackie walks through why AI can absolutely help us scale practice, create interactions faster, and give learners more chances to rehearse, reflect, and try again. But Jackie also draws a hard line: a conversation is not automatically learning, and “AI tutor” does not automatically mean “good feedback.” We break down what never changes in learning design, including clear goals, realistic scenarios, feedback tied to a standard, and a safe path back to real-world application. Then we get practical with four predictable risk areas (inaccurate feedback, generic responses, overtrust, and sensitive use cases like HR, legal, medical, mental health, or private data). From there, I share where AI is usually a best fit, plus five design requirements you can use to build better guardrails, including human escalation. You’ll also get a simple decision framework I call the Help Fit test: Helpful, Evidence-based, Low risk, Protected, along with a concrete customer service role play example to show what “good architecture” looks like. If you want to use AI in training responsibly, this is your roadmap. Subscribe, share the episode with a fellow designer, and leave a review if the framework helps you design smarter practice. 🔗 Episode Links Please check out the resource mentioned in the episode. Enjoy! AI Practice Compass [https://view.genially.com/6a00fe3251627ed38f182507]  Send Jackie a Text [https://www.buzzsprout.com/2146456/fan_mail/new] Join PodMatch! [https://www.joinpodmatch.com/designingwithlove] Use the link to join PodMatch, a place for hosts and guests to connect. Disclaimer: This post contains affiliate links. If you make a purchase, I may receive a commission at no extra cost to you. Support the show [https://www.buzzsprout.com/2146456/support] 💟 Designing with Love + allows you to support the show by keeping the mic on and the ideas flowing. Click on the link above to provide your support. ☕ Buy Me a Coffee [https://buymeacoffee.com/designingwithlove] is another way you can support the show, either as a one-time gift or through a monthly subscription.  🗣️ Want to be a guest on Designing with Love? Send Jackie Pelegrin a message on PodMatch, here: Be a guest on the show [https://www.podmatch.com/hostdetailpreview/designingwithlove] 🌐 Check out the show's website here: Designing with Love [https://www.designingwithloveblog.com/] 📱 Send a text to the show by clicking the Send Jackie a Text link above.  👍🏼 Please make sure to like and share this episode with others. Here's to great learning!

Kommentarer

0

Vær den første til at kommentere

Tilmeld dig nu og bliv en del af Designing with Love-fællesskabet!

Kom i gang

1 måned kun 9 kr.

Derefter 99 kr. / måned · Opsig når som helst.

  • Podcasts kun på Podimo
  • 20 lydbogstimer pr. måned
  • Gratis podcasts

Alle episoder

124 episoder

episode AI Tutors, Coaches, and Practice Bots: When They Help and When They Don't cover

AI Tutors, Coaches, and Practice Bots: When They Help and When They Don't

A bot can sound warm, responsive, and confident, and still teach the wrong thing. That’s the tension we dig into as we explore AI tutors, AI coaches, and practice bots through the lens that matters most to instructional designers: practice design that improves real performance, not just chat transcripts. Jackie walks through why AI can absolutely help us scale practice, create interactions faster, and give learners more chances to rehearse, reflect, and try again. But Jackie also draws a hard line: a conversation is not automatically learning, and “AI tutor” does not automatically mean “good feedback.” We break down what never changes in learning design, including clear goals, realistic scenarios, feedback tied to a standard, and a safe path back to real-world application. Then we get practical with four predictable risk areas (inaccurate feedback, generic responses, overtrust, and sensitive use cases like HR, legal, medical, mental health, or private data). From there, I share where AI is usually a best fit, plus five design requirements you can use to build better guardrails, including human escalation. You’ll also get a simple decision framework I call the Help Fit test: Helpful, Evidence-based, Low risk, Protected, along with a concrete customer service role play example to show what “good architecture” looks like. If you want to use AI in training responsibly, this is your roadmap. Subscribe, share the episode with a fellow designer, and leave a review if the framework helps you design smarter practice. 🔗 Episode Links Please check out the resource mentioned in the episode. Enjoy! AI Practice Compass [https://view.genially.com/6a00fe3251627ed38f182507]  Send Jackie a Text [https://www.buzzsprout.com/2146456/fan_mail/new] Join PodMatch! [https://www.joinpodmatch.com/designingwithlove] Use the link to join PodMatch, a place for hosts and guests to connect. Disclaimer: This post contains affiliate links. If you make a purchase, I may receive a commission at no extra cost to you. Support the show [https://www.buzzsprout.com/2146456/support] 💟 Designing with Love + allows you to support the show by keeping the mic on and the ideas flowing. Click on the link above to provide your support. ☕ Buy Me a Coffee [https://buymeacoffee.com/designingwithlove] is another way you can support the show, either as a one-time gift or through a monthly subscription.  🗣️ Want to be a guest on Designing with Love? Send Jackie Pelegrin a message on PodMatch, here: Be a guest on the show [https://www.podmatch.com/hostdetailpreview/designingwithlove] 🌐 Check out the show's website here: Designing with Love [https://www.designingwithloveblog.com/] 📱 Send a text to the show by clicking the Send Jackie a Text link above.  👍🏼 Please make sure to like and share this episode with others. Here's to great learning!

I går13 min
episode Designing Learning That Actually Changes Behavior with Dr. Steven Linley cover

Designing Learning That Actually Changes Behavior with Dr. Steven Linley

Training that doesn’t change behavior is just content, and content alone won’t move a business metric. Jackie sat down with Dr. Steven Linley, a learning strategist and adult learning expert, to unpack how to turn “we need training” into performance that lasts. From the first stakeholder request to the final coaching touchpoint, Steven shows how to investigate like a detective, separate red herrings from root causes, and design solutions that fit the real constraints of work. If you’re ready to move from order taker to strategic partner, this episode gives you the questions to ask, the levers to pull, and the system to build. Subscribe, share with a teammate, and leave a review to tell us: what’s the one shift you’ll make this week to turn learning into performance? 🔗 Website and Social Links: Please visit Steven Linley’s website and social media links below. Steven Linley’s Website [https://www.fokusgroup.io/] Steven’s LinkedIn Page [https://www.linkedin.com/company/fokus-group-led/] 📢 Call-to-Action: This call-to-action link directs listeners to the FōKUS website [https://www.fokusgroup.io/], where they can explore practical learning solutions designed to improve performance—not just deliver training. The site is especially valuable for leaders and teams looking to strengthen customer service, frontline performance, and professional development through well-designed, human-centered learning systems.  Send Jackie a Text [https://www.buzzsprout.com/2146456/fan_mail/new] Join PodMatch! [https://www.joinpodmatch.com/designingwithlove] Use the link to join PodMatch, a place for hosts and guests to connect. Disclaimer: This post contains affiliate links. If you make a purchase, I may receive a commission at no extra cost to you. Support the show [https://www.buzzsprout.com/2146456/support] 💟 Designing with Love + allows you to support the show by keeping the mic on and the ideas flowing. Click on the link above to provide your support. ☕ Buy Me a Coffee [https://buymeacoffee.com/designingwithlove] is another way you can support the show, either as a one-time gift or through a monthly subscription.  🗣️ Want to be a guest on Designing with Love? Send Jackie Pelegrin a message on PodMatch, here: Be a guest on the show [https://www.podmatch.com/hostdetailpreview/designingwithlove] 🌐 Check out the show's website here: Designing with Love [https://www.designingwithloveblog.com/] 📱 Send a text to the show by clicking the Send Jackie a Text link above.  👍🏼 Please make sure to like and share this episode with others. Here's to great learning!

7. juni 202641 min
episode From Content Creator to Learning Architect: The AI-Era Shift cover

From Content Creator to Learning Architect: The AI-Era Shift

AI can generate outlines, scripts, quizzes, scenarios, and slide drafts in minutes. That sounds like freedom, until you realize the real danger is volume: more content, more assets, more “resources” that don’t actually change what learners do. We’re making the case for a different kind of value in the AI era of instructional design and learning experience design: becoming the learning architect who decides what belongs, what gets left out, and what actually supports performance.  We break down the content factory trap, the pattern where deliverables become the goal and practice gets squeezed out. When everything feels important, learners get overwhelmed and the experience loses clarity. We reconnect to what still matters no matter how fast generative AI gets: structure, a clear learning path, meaningful practice, timely feedback, and support after training ends.  You’ll get three core architecture decisions to use on every project: the performance decision (what learners must be able to do), the experience decision (how they practice and build confidence), and the support decision (what helps them apply skills later). Then we make it practical with a simple learning architecture map you can use before you build anything: Outcome, Experience, Support, Evidence. If you want AI-assisted course design to lead to better results, not just faster production, this framework keeps you grounded in strategy.  Subscribe, share with a fellow instructional designer, and leave a review to help more designers build learning that works. 🔗 Episode Links Please check out the resource mentioned in the episode. Enjoy! Learning Architect Compass [https://view.genially.com/6a00edf368e76362e7241230]  Send Jackie a Text [https://www.buzzsprout.com/2146456/fan_mail/new] Join PodMatch! [https://www.joinpodmatch.com/designingwithlove] Use the link to join PodMatch, a place for hosts and guests to connect. Disclaimer: This post contains affiliate links. If you make a purchase, I may receive a commission at no extra cost to you. Support the show [https://www.buzzsprout.com/2146456/support] 💟 Designing with Love + allows you to support the show by keeping the mic on and the ideas flowing. Click on the link above to provide your support. ☕ Buy Me a Coffee [https://buymeacoffee.com/designingwithlove] is another way you can support the show, either as a one-time gift or through a monthly subscription.  🗣️ Want to be a guest on Designing with Love? Send Jackie Pelegrin a message on PodMatch, here: Be a guest on the show [https://www.podmatch.com/hostdetailpreview/designingwithlove] 🌐 Check out the show's website here: Designing with Love [https://www.designingwithloveblog.com/] 📱 Send a text to the show by clicking the Send Jackie a Text link above.  👍🏼 Please make sure to like and share this episode with others. Here's to great learning!

3. juni 202612 min
episode Assess What Matters In An AI World with Hamza Sami cover

Assess What Matters In An AI World with Hamza Sami

What if assessment made thinking visible and turned AI into a learning partner instead of a shortcut? Jackie sat down with Hamza Sami to close our series by unpacking practical ways to design for real understanding—where reasoning, judgment, and context take center stage. We start by reframing purpose: assessment is for learning, not just measurement. That lens leads to formative moves that build trust and invite responsible AI use. Hamza breaks down how to set clear AI norms, teach limits and risks, and require transparent acknowledgment with screenshots, links, and prompt logs. From there, we get tactical: in-class AI critiques, compare-and-verify exercises, and concise reflections that reveal what the model did well, where it failed, and how students adapted outputs to their goals. For summative checks, we spotlight formats that hold up with AI in play: presentations with Q&A, reflective blogs, portfolios of evidence, capstones grounded in authentic problems, and open-ended scenarios that demand justification. Hamza also shares three quick wins any instructor can deploy next term—explain-your-thinking checkpoints, brief peer feedback moments, and explicit AI process citations—plus one culture shift: stop letting assessment end at submission and bring learning to life through dialogue. Subscribe, share this episode with a colleague who’s redesigning a course, and leave a review to help more educators find the show. What’s the first assessment move you’ll try next? 🔗 Website and Social Links: Please visit Hamza Sami’s website and social media links below. Hamza Sami’s Website [https://urldefense.com/v3/__https:/www.behance.net/hamzasami__;!!MPeWGmKwT93l!AdfdCB2UvEXzTxE6VeL68T8-2Vi3xVGMW-oegC39pV8Z8Z8LGK4sbqGxtYj7n1cXpWxRiZM08vwn9TNQKdof8yw$] Hamza’s LinkedIn Page [https://www.linkedin.com/in/hamzasami/] 📢 Call-to-Action: Need a hand bringing your learning ideas to life, whether curriculum design, instructional strategy, or even creative production? Reach out on LinkedIn, and let’s explore how we can make learning more engaging together. Send Jackie a Text [https://www.buzzsprout.com/2146456/fan_mail/new] Join PodMatch! [https://www.joinpodmatch.com/designingwithlove] Use the link to join PodMatch, a place for hosts and guests to connect. Disclaimer: This post contains affiliate links. If you make a purchase, I may receive a commission at no extra cost to you. Support the show [https://www.buzzsprout.com/2146456/support] 💟 Designing with Love + allows you to support the show by keeping the mic on and the ideas flowing. Click on the link above to provide your support. ☕ Buy Me a Coffee [https://buymeacoffee.com/designingwithlove] is another way you can support the show, either as a one-time gift or through a monthly subscription.  🗣️ Want to be a guest on Designing with Love? Send Jackie Pelegrin a message on PodMatch, here: Be a guest on the show [https://www.podmatch.com/hostdetailpreview/designingwithlove] 🌐 Check out the show's website here: Designing with Love [https://www.designingwithloveblog.com/] 📱 Send a text to the show by clicking the Send Jackie a Text link above.  👍🏼 Please make sure to like and share this episode with others. Here's to great learning!

31. maj 202647 min
episode Your ID Knowledge Vault: How to Stay Consistent When AI Is Fast cover

Your ID Knowledge Vault: How to Stay Consistent When AI Is Fast

AI can help you generate course content in minutes, but if you’ve ever looked at the output and thought, “This doesn’t sound like me,” you already know the hidden cost: inconsistency. When tone changes, terminology drifts, and structure varies across modules, your work stops feeling recognizable and trustworthy even if the content is technically correct. We walk through a simple fix that doesn’t require a massive system or 100 documents: an Instructional Design Knowledge Vault. I explain why the real risk isn’t AI, it’s using AI without a home base for your voice, your standards, and your repeatable design patterns. We name the three predictable problems that show up without a vault (brand drift, duplicate work, and the rework spiral), then build a practical four-folder setup you can keep in whatever tool you already use: Voice, Standards, Patterns, and Proof. You’ll leave with a starter list of assets you can create quickly, including a five-bullet voice guide, a QA scan focused on facts, fairness, and voice, prompt templates for spec, critique, and variations, plus a single gold-standard example to guide future drafts. Most importantly, we share one rule that turns “random output” into consistent output: every time you prompt, attach one vault item. If you want AI speed without losing your quality bar, this is the workflow to try next. Follow or subscribe for more practical, human-centered instructional design strategies, and if this helped, share it with a fellow designer or leave a review so more people can find the show. 🔗 Episode Links Please check out the resource mentioned in the episode. Enjoy! ID Knowledge Vault Compass [https://view.genially.com/6a00d11ca0c07669904302a8]  Send Jackie a Text [https://www.buzzsprout.com/2146456/fan_mail/new] Join PodMatch! [https://www.joinpodmatch.com/designingwithlove] Use the link to join PodMatch, a place for hosts and guests to connect. Disclaimer: This post contains affiliate links. If you make a purchase, I may receive a commission at no extra cost to you. Support the show [https://www.buzzsprout.com/2146456/support] 💟 Designing with Love + allows you to support the show by keeping the mic on and the ideas flowing. Click on the link above to provide your support. ☕ Buy Me a Coffee [https://buymeacoffee.com/designingwithlove] is another way you can support the show, either as a one-time gift or through a monthly subscription.  🗣️ Want to be a guest on Designing with Love? Send Jackie Pelegrin a message on PodMatch, here: Be a guest on the show [https://www.podmatch.com/hostdetailpreview/designingwithlove] 🌐 Check out the show's website here: Designing with Love [https://www.designingwithloveblog.com/] 📱 Send a text to the show by clicking the Send Jackie a Text link above.  👍🏼 Please make sure to like and share this episode with others. Here's to great learning!

27. maj 20269 min