As a side note, I also just realized that Nano Banana (the image model in Gemini) can make mockups. Here’s the very simple prompt I used:

Make a photorealistic mockup of [the screenshot] on a laptop on a desk in a cozy home office with #adc3df walls. A cup of coffee is on the desk. A bookshelf is visible in the background. No books on the desk or people or plants. 

Original data error spotting activity

I created a similar activity for an introductory data analysis course relating to a learning objective on spotting common errors in data. I’ve changed the questions, data, and other details so I’m not sharing any of my client’s actual information. You can try the activity yourself below (although the spreadsheet data means this activity works best on a larger screen).

Embedding the activity directly in my blog post has been challenging because of the variable height of the activity (plus, embedded html doesn’t play nicely with email). You can view both versions as Rise blocks or open the activity in a new tab.

This activity isn’t the flashiest use of AI, but it’s useful. Users get a way to look at sample data and practice finding errors, which is exactly the skill I needed to support. It’s fun to vibe code things like arcade games, but most of the time workplace training benefits from activities that look like people’s work.

How I created this activity

I created this vibe coded activity in Claude. As I observed after Jeff Batt’s vibe coding session at Training 2026 earlier this year, vibe coding really can be done by anyone now. The key skill is being able to clearly describe what you want and to provide the AI with specific feedback to iterate and improve. Your skills in describing interaction and giving clear directions matter more than your technical skills.

The good news is that instructional designers already have those skills, especially those of us who have experience writing storyboards to pass off to an elearning developer to build. When I started as an ID, I created PowerPoint storyboards and handed them off to our team of Flash developers. That meant I had to very explicitly note how I wanted interactions to work. If I didn’t write clear directions, then the Flash developers couldn’t understand what I wanted and build it. Vibe coding feels very similar, except now it’s the AI doing the role of that custom development.

My prompt to create the activity wasn’t technical; I focused on the content and interactivity.

I'm working on a course on data analysis 101 for [the role in my client's audience]. I need some examples of data errors for some "spot the error" questions in each of these categories.

Create a web-based interactive activity for each of these types of errors. Include some sample data that would be related to [client's industry--changed for the example above]. For each sample, users will click on the mistake in the data. Then, the feedback will confirm if they found the error (or errors) and tell them what type of error this was.

The activity must be a self-contained HTML file with inline CSS and Javascript so I can embed it in an Articulate Rise block.

My prompt continued by listing the four types of data errors and some of the tips from the course content that learners receive prior to this activity.

Iteration and testing

After I got the results, I did some iteration and testing.

• Changed the question labels to numbers; the original version said what kind of error it was, which made the activity too easy.
• Adjusted the instructions to avoid giving too many hints.
• Added follow-up questions about the type of error.
• Changed the colors to match my client’s branding.

All of that iteration happened through ordinary chat with Claude. Obviously, I could build something like this in Storyline, and I have done so in the past. However, this kind of activity is really tedious to build with multiple hotspots or rectangles and triggers. This took about 45 minutes to create, test, and iterate in Claude. That’s saving a significant amount of time.

Improving the accessibility

Unfortunately, I screwed up. I didn’t include any requirements for accessibility in my initial prompting and testing. So, I asked Claude if this activity would meet WCAG accessibility standards and if a screen reader user could use it. Claude replied, “Honest answer: as written, this activity does not fully meet WCAG 2.1 AA, and a screen reader or keyboard-only user could not complete Part 1.” So, I went through some more iteration to improve the accessibility. This still may not be fully accessible for all users, but it’s much more accessible.

This is where my big insight comes in: I really should have prompted for accessibility right from the start. I created extra work and testing by not asking for it in the first place. Fortunately, even though I did mess up by not planning for accessibility at the beginning, it didn’t take Claude that long to rework everything. Claude can build accessibility features, but you have to ask for them.

Accessibility problems and fixes

Claude helpfully suggested making a table with a side-by-side comparison of the problems and everything needed. While I write my blog posts myself with my human brain, I copied this table straight out of my chat.

More accessible version

You can try the vibe coded activity with improved accessibility below. From a user perspective, if you use a mouse, not much looks different. But if you click one question and then tab through the activity, you’ll see that you can now access each cell in the table with a keyboard. You can also see the difference in how the activity previously conveyed feedback through color alone previously, but now also marks answers with an X or check mark.

More on AI

Check out all of my posts about using AI for instructional design for more posts where I show my work (the good and the bad) like this.

---

Upcoming events

Crafting Choices That Challenge: Designing Scenarios for Authentic Practice. July 15 at 12PM Pacific / 3PM Eastern. Register for this free webinar hosted by Training Mag Network.

Scenario-Based eLearning Design Certificate: Strategies to Engage, Motivate, and Transfer Skills. 3 online sessions starting August 5. Save $150 with Early Bird Registration.

Creating High-Quality AI Media: Prompting, Scripts, and Consistency. October 8. Register for the ATD AI Intensive.

DevLearn

Branch Smarter, Not Harder: Naming Conventions, Variables, and Bottlenecks in Storyline

• November 3, 2026 at 9:20 AM
• CE01: Storyline & Rise Pro Lab

Cohesive by Design: Generate Consistent AI Icons, Illustrations & Characters

• November 4, 2026 at 10:00 AM
• BYOD (Bring Your Own Device)

The post How I Made an Accessible Vibe Coded Activity appeared first on Experiencing Elearning.

This post contains affiliate links. It doesn’t cost you anything extra, but I earn a small percentage when you buy through these links.

Generalist or specialist

Generalists do a little bit of everything. This is pretty common in instructional design and elearning roles, especially in smaller organizations. Generalists have a broad set of skills. You can have a lot of variety in your work when you’re a generalist.

True specialists have deep skills in one or two areas, but tend to have a weaker base of other related skills. Elearning developers are often specialists; they work best when someone else provides them with a storyboard or script that they can be creative in building and development. AI opens up the possibilities for new specialties in AI media creation or working with agents. Larger teams are more likely to hire specialists who focus on a particular skill or part of the process.

Unicorn or comb

What employers ask for in job openings is often unicorns: people who have deep skills in all the areas. In practice, real unicorns aren’t very common. The quote below is about UX design, but this framework applies to instructional design as well.

Aim towards unicorn, realistically end up as a beautifully misshapen comb

The reality is that to be a unicorn is a worthy goal, but most people won’t get there, and that’s fine! First, find something you’re good at and which contributes towards the field you’re interested in, then you have at least something to fall back on and can bring value to a broad range of teams.

Sophie O’Kelley in Generalist versus specialist? How about a misshapen comb?

Rather than being a unicorn, it’s more common to be a bit of a “misshapen comb.” That means you have the broad base of skills that give you a little knowledge about many topics and help you interact with other specialists. You also have a couple of deeper skills, maybe at different depths. An elearning developer might have deep skills in Storyline, plus moderate skills in AI media creation, UX, graphic design, and a little Javascript, along with a foundation of instructional design principles.

The idea of the comb is related to Cammy Bean’s explanation of broad and deep T-shaped skills in her book, The Accidental Instructional Designer.

It’s effectively a hybrid: you have that broad base of skills as a generalist that makes you an asset for working across teams. You also have some specializations of various depths. That combination of specialist skills are what set you apart from every other generalist out there.

Good organizational fit

Generalists

Generalists have traditionally lots of options for jobs in ID. They can contribute to a team or work as a department of one (fairly common in L&D). Being a generalist is how people often start out in the field of instructional design because you get experience across a broad range of skills.

But being a generalist also can be challenging because you don’t have as clear of a way to stand out in the job market. Your skills may need to look more like the “misshapen comb” from the article I quoted earlier; you have a broad base of generalist skills plus several moderate skills and one or two deep specialties.

Everything I hear from folks looking for jobs right now is that the market is rough, and the competition is hard. Even people with years of instructional design experience and credentials are having a hard time. Entry level generalist positions in ID were hard to come by even before AI, and it seems to be getting harder now as AI redefines roles and other pressures increase competition for jobs.

Specialists

The bad news is that means you may need some sort of specialization to stand out. The good news is that AI is so new that even if you don’t have a ton of experience in ID that almost nobody has more than a few years head start building their skills in AI. There are opportunities to create new specialties for yourself as differentiators.

For example, five years ago, AI image generation didn’t exist. But now I’m not only using AI images for my work, I’m getting paid to teach people about AI images too. It’s taken a lot of work and sharing my expertise for free to get to this point, but I created a specialty for myself. That specialty means I fit with some organizations better than others.

If you’re a true specialist who wants to work in only one area, you may need to be more discerning in searching for jobs or projects that align with your specialty. Larger companies are more likely to have larger teams where different members have different specialties. Elearning vendors, the companies that specialize in creating elearning for other organizations, are other good places to find more specialized roles.

In the past, elearning development specialists seemed to have an easier time finding work, especially as freelancers. Organizations were always looking for developers who can build elearning, even if they have an internal team of instructional designers. Development is the easiest part of the process to outsource to a freelancer or consultant. That might still be true today, but it’s unclear to me right now how that will shift as organizations move away from a short list of traditional authoring tools and into a wider range of authoring and AI tools. Developers with skills using AI tools may have an advantage, at least in the short run.

I know people who want to focus primarily on writing and stay away from development or technical work. Job searching can be a trickier if you want to focus primarily on strategy, analysis, design, and storyboarding. You have to look for those larger teams where roles are specialized, or at least organizations who regularly outsource development work. Those roles are increasingly challenging to find though. Jobs focused on a narrower range of tasks are more at risk from AI. Shifting yourself closer to the business problem and into more performance consulting may be one way to avoid working with development.

Hybrid generalist/specialist

A stronger position is that “misshapen comb” or hybrid generalist/specialist. Combining a broad range of generalist skills means you’re adaptable and can do a bit of everything. Generalists who do many different tasks are harder to replace with AI or automation, and they’re good at combining skills. But there are so many generalists out there that it’s hard to get a job if your differentiator is something like, “I used to be a teacher, and now I’m an ID.”

Pick one skill you’re interested in learning more about: learning science, some aspect of AI, video, UX, scenarios, games, customer enablement, onboarding, etc. You don’t have to learn everything and have deep skills in all of it, but growing your skills in specific areas makes it easier to find roles that use those skills.

Questions to ask

Many people treat job interviews as a one-sided evaluation of your skills, instead of a two-way review for good fit. When you interview, asking questions helps you figure out the role and the culture.

Ask questions like these to learn more about the position:

• What does a typical day or week look like in this position?
• Thinking about the people who have been really successful in this role in the past, what sets them apart from people who are “just OK” in this role? (If they say how pretty they make PowerPoint, but you know you want to focus on performance consulting, you know it’s not a good fit.)
• What’s the most challenging part of this position?
• Tell me about the roles in the rest of the team. Who else would I be working with?
• How do you determine which projects get developed and what projects have priority?
• How do you use AI in your instructional design processes?

If you ask those sorts of questions, you should get a better feel for the job and the organizational culture. Even interviewing with an elearning vendor as a subcontractor, you could ask some of these questions. The answers will help you figure out how much is just order taking from clients and how much is consulting.

More on Instructional Design Careers

Check out all of my posts on instructional design careers, including these:

• Is instructional design the right career?
• Teacher to Instructional Designer: Interview Questions
• Instructional Design Hourly Rates and Salary
• Instructional Design Portfolio Resources
• Instructional Design Careers and Freelancing Presentations and Podcasts

Originally published 11/19/2019. Updated 6/18/2026.

The post Finding a Good Fit in Instructional Design appeared first on Experiencing Elearning.

Obviously, I’m not a lawyer, and nothing I say here is legal advice. This is my understanding of fair use and copyright. Check with your own attorney if you have legal questions.

My goal here isn’t to convince you to use AI images if you’ve already decided it goes against your ethical values. If that’s you, I respect that, and I’m not here to change your mind. I’m writing for those who are using AI images and want to do so carefully.

Training data

I know lots of folks who oppose AI, and AI images particularly, because of the training data. I think it’s totally reasonable to use Adobe Firefly for image generation because it was trained only on licensed stock and public domain images. Because they have permission for everything they used to train their model, that’s the safest option. That’s an easily defensible ethical stance.

Most AI image generation models were trained on a much broader set of data. That includes photos and art that are available online but not in the public domain. Just like LLMs were trained on copyrighted text, image generation models were trained on copyrighted images. That means AI companies are relying on a fair use argument.

I’ve seen a range of legal opinions on AI training. This is clearly an area where the law is still evolving. So far, the decisions support training models on copyrighted materials as long as they were legally obtained. Anthropic is paying a settlement for using pirated books to train on, but the books they legally bought were allowed under fair use. That’s where I generally fall too; as long as the training data wasn’t pirated, training on a broad range of publicly available images is fair use.

I don’t follow the legal landscape in the EU as much as I do in the US. However, it looks like the EU AI Act will require more transparency on data sources. (Any readers with more insight on the EU, feel free to comment and correct me if I’m wrong about that.) I hope we do get more transparency on training data so people can make informed choices about which tools they use or avoid.

Transformative use

Fair use is the US doctrine that allows some use of copyrighted content under certain circumstances. Fair use is very “squishy” though; it’s not a clear cut rule. It’s not as simple as saying there’s a clear line based on any single factor. You have to weigh four factors, and judges sometimes decide individual cases differently even with similar circumstances.

If someone tells you that something is definitely not fair use because it uses the whole source while ignoring the other three factors, be very skeptical. That’s not really how fair use works. Yes, the amount of the source material used matters, but that’s only part of the question.

One of the other four factors is the purpose and character of your use of source content. That’s why parody and satire are often allowed even of copyrighted or recognizable sources; that’s a protected purpose that constitutes fair use. The judge in the Anthropic case ruled that AI training is a transformative use; it transforms the original into something else. Because the character of the use is transformative, it’s fair use.

For my own work, that ruling means I’m less concerned with the input for AI and more concerned with the output. Because of course you can make things with AI that are copies (or very close to copies) and aren’t really transformative. Just because an AI image model lets you generate something obviously based on Star Wars or the Simpsons doesn’t make it transformative enough to be considered fair use. And regardless of the legal landscape, transformative use is more ethically defensible.

Don’t emulate living artists

One of my own guidelines for using AI image generation is that I don’t use it to emulate the style of any single living artist. You want to make a version of Starry Night with a cat in it? Cool, go for it! Van Gogh isn’t alive to be affected by it. But any artist who is alive now, who you (at least hypothetically) could hire to create something? Nope. While it was trendy to do Studio Ghibli style remixes for a while, I opted not to participate in that.

Copying the style of a living artist potentially affects the market for that person’s art—and that’s another of the four factors weighing against fair use. Plus, that use doesn’t feel right to me. Copying a single style doesn’t feel transformational. Plus, why use these tools to simply copy when they’re capable of more?

Instead of emulating a specific artist, use AI to develop a new style. That’s one of the reasons I still use Midjourney even though other image generators are now better at consistency and prompt adherence. Midjourney is the best for creating new styles and mashing up elements of multiple images together.

It’s like the old academic joke: if you copy from one source, it’s plagiarism, but if you copy from a bunch of sources, it’s research. AI image styles feel similar. If you copy from one artist, it’s plagiarism. If you copy from a bunch of artists, it’s transformative and fair use.

AI images can’t be copyrighted

Like anything else produced by AI, AI-generated images can’t be protected by copyright unless a human makes substantial changes to it. Work that is primarily guided by the decisions of a human but is “AI-assisted” may be copyrighted. Read AI and Copyright for more details and examples of what kinds of edits and decisions make something “AI-assisted” rather than “AI-generated” and an explanation of the March 2026 US Supreme Court decision.

In my opinion, that doesn’t matter too much for images within training. If you were using stock images previously, you didn’t own the copyright to those images either.

While there are multiple AI tools for creating business logos, that feels really risky to me. Even if you’re a small business and don’t plan on trademarking your logo, wouldn’t you like the option of doing so in the future? Pay for a human to create your logo so you can protect that IP.

The safest tool related to IP claims is Adobe Firefly, which also gives you indemnity against IP claims with paid plans.

As a side note, this is one of the issues with using AI to generate course content too, especially courses to sell. Someone can duplicate your content and sell it, and you might not have much recourse. This is an area of the law to keep watching, since future court decisions or legislation may change it.

Commercial use of AI images

Check the ToS for your AI image tool. Most free tools for AI images don’t license your results for commercial use. You generally have to use at least a minimal paid plan to get commercial use rights.

Don’t use images of real people without permission

Don’t use images of real people as reference images for AI unless you have explicit permission to do so. That’s not just for cases of deepfakes or sexually explicit images; that goes for everything.

I see way too many examples on LinkedIn where people have done “fun” transformations of people’s photos. Most of that isn’t malicious. It’s stuff like people making composite group images or turning workshop participant images into superheroes or whatever. But people should have the right to not have their images used that way, where your identity is clear.

Plus, even with good intentions, sometimes the images are problematic. I remember seeing someone’s AI-generated image of people they were looking forward to having a beer with…except I noted multiple people who don’t drink shown in that image. I shouldn’t have to explain that you shouldn’t generate images of a Jewish person eating bacon or a Mormon drinking alcohol, but apparently that’s a thing that need to be explicitly mentioned.

Even if you have stock images of real people, you may not have permission to use those images as references in AI. If you need character images, just use AI to generate your reference image and work from there.

Environmental impacts of AI images

The water use for AI data centers gets a lot of attention, but every time I actually dig into the sources and data, I see that the water use is relatively minimal. I don’t want any AI data centers drawing water from the Colorado River where water use is already such a problem. However, I think data centers can be built in places where water is more available, and they can be built and managed in ways that reduce their impact.

Electricity use is potentially a problem, especially at scale. AI models are much more efficient than they used to be, so they don’t use as much energy as early models (at least for regular prompting). I’d love to see more investment in renewable energy in the US. Where I live in North Carolina, data centers don’t pay taxes on electricity. I’d like that to change; I think data centers shouldn’t be incentivized so much that it’s hard to track the real costs.

You can also reduce your energy use by using cheaper models when possible. I often use Midjourney’s draft mode if I know I need a lot of iteration to fine tune something, partly because that reduces my energy use.

It’s important to keep the water and electricity use of AI in perspective with other tasks though. Check out the What Uses More? calculator to compare the energy and water use of different tasks.

For example, would you think twice about the water and energy use for a one-hour Zoom call with 10 people? Probably not.

Guess how many images could you generate with AI for the same resource use as that one-hour Zoom call?

According to the calculator, that one-hour Zoom call is equivalent to generating 1000 images. I generate a lot of AI images, but even I have only generated about 1500 images in Midjourney over the course of two years. Are you going to generate 1000 images in a year? How many Zoom calls will you have in that same year?

Transparency in AI use

One question that comes up often when I present about AI images is about transparency and labeling AI images. This is an interesting question because I don’t think there’s a clear answer that fits every situation. I had a client project where I clearly labeled everything I generated with AI but was asked to remove the disclosures. I can’t find the source right now, but I remember reading somewhere that Americans were generally less worried about having AI labeled, but Europeans (especially Germans) value transparency more. Cultural differences play a role here, as does organizational culture.

I used to label every AI-generated image here on my blog, but these days I admit that I figure people mostly know even without the label. But maybe I should be more explicit? From an ethical standpoint, transparency feels like the more defensible position.

Counter bias in AI images

AI image generators tend to exaggerate stereotypes. If you don’t actively work to counter bias in AI images, it can be worse than stock images. The good news is that AI images can give us even better, more inclusive representation than stock images. We’ve had the ability to create images of nonbinary characters and people of varied racial, ethnic, and national backgrounds for a while. We’re now finally getting to the point where disability representation is better. Inclusive images is one area where I think AI can give us better outcomes than stock photos. Using AI for social good is an ethical consideration too.

Review AI images

Just like you need to review AI-generated text for accuracy, review your AI-generated images. Beyond obvious AI errors, check the background objects, watch out for bias, and verify the results. If you always accept the first image result without refining your prompt or editing the result, that’s a problem. Your human judgement is more important than ever.

Further reading on AI ethics and legal considerations

• Is Any AI Use Ethical?
• Creating AI Art Responsibly: How to Navigate AI Legal Risks
• AI and Branding 2026: Copyright Risks for Content Creators
• Using AI Responsibly For Image Generation: Ethics And Transparency Matter
• The Ethics of AI Generated Images
• Paradox Learning Resources, including an AI Ethics Rubric
• Fairly Trained (certifies consent and licensing in training data)
• Thaura, ethical AI

What do you think?

What are your ethical guidelines for AI image generation? Do you prefer to see AI images transparently disclosed?

I expect a fair amount of disagreement on this. I’ve had some great discussions about this though, even with people who don’t agree. I’m hoping for some nuanced conversations about balancing our ethical considerations. Let me know your thoughts in the comments.

The post Considerations for Using AI Images Ethically appeared first on Experiencing Elearning.

Automation

After Automation | Every

If you’ve ever spent time cleaning up an AI-generated draft, you’ll get the initial point here: AI creates more work for humans, not less. But this article digs deeper into how working with AI, especially AI agents, changes the nature of work. Employees at this author’s company spend more time directing agents (deciding the goal and what “good” looks like) and on judging the results of AI. The easy work that AI can do gets commoditized, but the hard work of taste and judgement become more important.

There’s no tipping point coming where things flip and the jobs are gone. The new reality is the opposite—the more we automate, the more expert human work there is to do. Here’s why: AI commoditizes the residue of human expertise—whatever can be made explicit enough to train on. That collapses the value of default model output and creates demand for what’s different. Demand for what’s different is demand for human experts, even as we approach artificial general intelligence (AGI).

Across both forms—coworker and embedded—the pattern is the same. Employee agents take over more of the stable, repeatable, well-framed layer of work. But there is a lot of work that still requires a human being in the loop. We’ve found over and over that for any kind of complex task, the best way to get great work is to have an AI and a human going back and forth in the same workspace.

In every example, the agent needs a human in order for the work to, well, work. Someone has to point it at the right thing, decide whether the output is good, catch the places where it is wrong, and turn the result into a real-life decision or process. The further away an agent gets from a human who is in charge of making sure it works well, the less well it works.

When work is abundant and looks alike everywhere, the work that doesn’t fit the pattern becomes the rare, valuable, and high-status thing(5).

This is why, in practice, AI does not eliminate expert human knowledge work. It dramatically increases the volume of work being done, and none of that work is differentiated or valuable unless a human being is involved.

—Dan Shipper

AI Brain Fry, Workslop and the Ironies of Automation

This is a long article, but worth spending some time to digest. One of the ways that AI changes the nature of work is by increasing the amount of time we spend in cognitively challenging tasks like evaluation. But human brains need breaks and variety.

What remains after automation is not a simplified role but an arbitrary residue of the most demanding, most ambiguous, and least supported work in the entire system. The human is not replaced. In other words, the human is paradoxically left with the hardest parts, and given almost no preparation for them.

Surveying nearly fifteen hundred full-time workers across industries, roles, and seniority levels, the researchers found that intensive oversight of AI tools was the single most mentally taxing form of engagement their participants described. Workers required to monitor AI agents closely reported fourteen percent more mental effort, twelve percent more mental fatigue, and nineteen percent greater information overload than those whose AI engagement was less demanding.

The final irony of automation, she wrote, is that the most successful automated systems, those with the rarest need for human intervention, are precisely the systems that require the greatest investment in human skill.

—Carl Hendrick

AI

AI and Copyright: What You Can and Can’t Do with AI-Generated Content (2026)

An overview of AI and copyright. Who owns it? When can you use it for commercial purposes? How can you protect your AI-assisted work?

Image AI prompts

A large collection of images generated in Nano Banana, ChatGPT, and Seedream plus their prompts. Seeing how others have prompted for images is helpful in figuring out what works and what doesn’t.

AI Effectiveness Rating

Christopher Lind’s simulation tool for rating effectiveness working with AI

Oboe

Ask this AI about a topic and get a lesson on it. This is for self study, and it seems potentially useful for basic topics where there’s a lot of high-quality publicly available information to draw from. You can generate flash cards or study guides too. This might be something that’s more useful for students in school, but I can see it for employees wanting to learn the basics of visual design, copywriting, etc.

Scenarios

New Scenario-Question Generator — Version 5 » Work-Learning Research

Will Thalheimer released an updated version of his template for generating scenario questions. This includes notes about differentiating people who know how to do a task successfully and those who don’t know. The template also includes an annotated example scenario.

Clinical Case Study: Diane

This is a short scenario built in 7Taps by IDLance to use as spaced repetition to reinforce prior training. One aspect I really liked in this was the interaction where you listen to two people’s arguments for different courses of action. Then, you have to decide which is the better way forward. This technique could help make binary choices feel more realistic in the context of a scenario, especially if the justification for the worse answer is plausible for how people think.

Visual inspiration

21 Ways To Get Visual Ideas

Connie Malamed recently published a significant update to her article with resources for getting visual ideas and inspiration. This includes so many links to visual resources across the 21 categories.

---

Upcoming events

Crafting Choices That Challenge: Designing Scenarios for Authentic Practice. July 15 at 12PM Pacific / 3PM Eastern. Register for this free webinar hosted by Training Mag Network.

Scenario-Based eLearning Design Certificate: Strategies to Engage, Motivate, and Transfer Skills. 3 online sessions starting August 5. Save $150 with Early Bird Registration.

Creating High-Quality AI Media: Prompting, Scripts, and Consistency. October 8. Register for the ATD AI Intensive.

DevLearn

Branch Smarter, Not Harder: Naming Conventions, Variables, and Bottlenecks in Storyline

• November 3, 2026 at 9:20 AM
• CE01: Storyline & Rise Pro Lab

Cohesive by Design: Generate Consistent AI Icons, Illustrations & Characters

• November 4, 2026 at 10:00 AM
• BYOD (Bring Your Own Device)

The post Automation, AI, Scenarios: ID Links 6/9/26 appeared first on Experiencing Elearning.

Example scenario with inconsistent perspective

Let’s look at an example. For this scenario, the objective is to help new managers conduct effective performance conversations with underperforming team members.

Evie is a first-time manager. Her team member Lincoln has missed three deadlines in the past month, and the quality of his recent deliverables has dropped. Evie has scheduled a one-on-one to address it.

With this setup, Evie is the protagonist. She’s the manager who the audience should identify with. That means Evie is the character who should be making decisions; the action should focus on her.

But look at the first decision point after Evie starts the meeting and asks about the missed deadlines.

What should Lincoln say next?
A. Promise to do better at meeting the deadlines moving forward.
B. Explain the problems he's had getting critical information from another team.
C. Ask for help prioritizing his responsibilities.

Instead of a scenario about how managers talk to their employees, suddenly this has become a scenario about how employees respond to criticism. The action is focused on the employee’s perspective; it no longer aligns with the role of the manager.

How to check alignment

When you review scenario decisions, you’re watching for alignment between the audience role, the skill being trained, and the decision points themselves.

• Does the protagonist of the scenario have a similar role to the audience’s role in their work?
• Does the decision point show the protagonist acting, or does it show another character’s actions?
• Does every decision point relate to the same character’s actions, or does it shift throughout the scenario?

Of course, it’s possible to create a branching scenario that shows multiple perspectives. Hana Feels is a great example of an interactive story where you make choices based on multiple characters who talk to Hana, who is ostensibly the main focus of the story. But you only guide Hana indirectly, trying to shape her emotional responses based on her conversations with others. A multi-character scenario like that requires a lot of planning though, and it’s more complicated to maintain a consistent narrative over the course of various paths. Most of the time, when I see this in training branching scenarios, it’s haphazard rather than well-planned.

How to fix the consistent perspective problem

Fixing the problem of inconsistent perspective requires stepping back to the planning stage. Every decision point should focus on what your protagonist is doing. The actions of other characters are the consequences and challenges that your protagonist responds to, rather than the decision points themselves.

When I start planning a branching scenario, one of my early steps is to draft an outline of the main actions. A simple numbered list like this helps make sure I’m on the right track with my focus before I get too far into prototyping or development. These actions reflect the “ideal path” someone might take if they followed the process perfectly and chose the best options every time.

For this example scenario on performance conversations, that outline of actions in the ideal path could look like this:

1. Opening: Open with a clear, specific statement of the issue.
2. Gather Info: Ask open-ended questions to understand the context.
3. Listen: Listen actively and acknowledge what she hears.
4. Set Expectations: Restate what success looks like to clarify expectations moving forward.
5. Action Plan: Clarify concrete next steps for follow-up and support.

With that outline guiding me, I know each of my decision points connects to those actions. Since the first step is the opening the conversation, I can diagnose the problem: my first decision point actually happens too late in the story, after Evie has already opened the conversation. I need to back up and put the decision point right at the beginning of their meeting.

What should Evie say?
A. "Your work hasn't been great lately. I know you can do better, so let's talk about how to get back on track."
B. "I can't believe you missed the deadline for the Q3 dashboard! That's the third deadline you've missed recently. This is a becoming real pattern, and it's a problem."
C. "In the past month, you've missed multiple deadlines: the Q3 dashboard, the Henderson report, and the Palmer project QA review."

Now the decision is clearly about the performance conversation and the actions Evie takes as a manager. Lincoln’s reactions are the consequences and effects of what Evie says. The next decision point will respond to Lincoln’s words. If he’s resistant to the feedback, then Evie needs to ask questions and actively listen to understand and work past that resistance. The scenario decisions remain focused on Evie’s actions though, not Lincoln’s responses to feedback. The scenario has a consistent perspective that reflects the learners’ experience.

What branching scenario challenges can I help you with?

Do you have a challenge writing or building branching scenarios? What would you like to see me write about in future posts? Let me know in the comments or by replying to this email.

---

Upcoming events

Crafting Choices That Challenge: Designing Scenarios for Authentic Practice. July 15 at 12PM Pacific / 3PM Eastern. Register for this free webinar hosted by Training Mag Network.

Scenario-Based eLearning Design Certificate: Strategies to Engage, Motivate, and Transfer Skills. 3 online sessions starting August 5. Save $150 with Early Bird Registration.

Creating High-Quality AI Media: Prompting, Scripts, and Consistency. October 8. Register for the ATD AI Intensive.

DevLearn

Branch Smarter, Not Harder: Naming Conventions, Variables, and Bottlenecks in Storyline

• November 3, 2026 at 9:20 AM
• CE01: Storyline & Rise Pro Lab

Cohesive by Design: Generate Consistent AI Icons, Illustrations & Characters

• November 4, 2026 at 10:00 AM
• BYOD (Bring Your Own Device)

The post Consistent Perspective in Branching Scenarios appeared first on Experiencing Elearning.

This update includes the podcasts I recorded in 2025-2026.

• Freelancing for IDs (ID Lounge Part 1, April 2026)
• Artificial Intelligence and Instructional Design (ID Lounge Part 2, April 2026)
• AI-Generated Images for Learning Content (IDIODC, February 2026 )
• The Future Isn’t Flashier—It’s Smarter (EdUp L&D Podcast, July 2025)

Jump right to the section you want:

• Scenarios and storytelling
• ID careers, freelancing, and consulting
• AI and instructional design
• Other topics: engaging learners, time and cost estimates, general ID

Podcasts on scenarios and storytelling

My Approach to Scenario Writing and Multimedia: Visual Lounge Podcast

My Approach to Scenario Writing and Multimedia: Visual Lounge Podcast

I shared tips for writing, using storytelling elements, and when to use multimedia to enhance your messaging. We explored why knowing your audience and context is the starting point of all successful content and how you can learn more about your audience’s challenges, scenarios, and goals (even when you don’t have direct access to the information).

The Power of Practice in Workplace Learning

The Power of Practice in Workplace Learning: Podcast Interview

Most workplace training stops at delivering information – but does that actually help people apply what they’ve learned? In this episode of The Learning+ Podcast, I joined the conversation to explore how scenario-based learning bridges the gap between knowledge and action. We discuss how practice and feedback build confidence, when branching scenarios are most effective, and how to get stakeholder buy-in. I share insights on how to make learning stick.

Scenario-Based Learning 101

Scenario-Based Learning 101: Animaker Interview

This is Scenario-Based Learning 101, serving primarily as an introduction to the topic for people who are just getting started with scenarios. I unpack some of the common misconceptions surrounding scenario-based learning, discuss the various types of scenarios, and offer practical advice on how to apply scenarios effectively in different areas of training.

Scenario-Based Software Training

Scenario-Based Software Training: Digital Adoption Show

What makes training truly effective? Is it about teaching “click here” steps, or is it about empowering learners to make real-world decisions with confidence? That was one of the core ideas in my interview on the Digital Adoption Show podcast from Whatfix. In this interview, I focused specifically on software training, addressing when scenario-based software training is the right approach and when you should use other training methods. Honestly, a lot of the time scenarios are overkill for software training, but they are helpful in some specific situations.

Scenario-Based Learning for Learning Impact

Scenario-Based Learning for Learning Impact

How do you increase learning impact through scenario-based learning? While no solution is right for every training situation, we know that providing opportunities for practice with feedback is critical for developing skills. Scenario-based learning is one way to provide that practice with feedback to help learners improve their performance. In this interview with Amit Garg on the L&D Go Beyond Podcast, I discuss scenario-based learning and how that can increase skill transfer to real-world environments.

Scenario-Based Learning Q&A with Devlin Peck

Scenario-Based Learning Q&A with Devlin Peck

In this interview, I discussed when to use scenarios, tips for working with SMEs and stakeholders, planning and designing branching scenarios, and tools and resources.

Scenario-Based Learning Experiences

Scenario-Based Learning Experiences Podcast Interview

For one of my first podcast interviews, Jacqueline Hutchinson of The Lounge podcast and I had a lovely conversation on scenario-based learning. She had some great questions about how to use scenarios and storytelling in learning experiences that led to a really fun chat. We talked about some “horror stories” and scenarios gone wrong, using scenarios to make compliance training not sleep-inducing, and options besides branching scenarios for incorporating storytelling.

MADE Podcast on Branching Scenarios

MADE Podcast on Branching Scenarios

My interview for the MADE podcast on branching scenarios: when to use them, challenges, tools, planning, and getting started. This is a short conversation (about 20 minutes), if you’re looking for a quick intro to scenario-based learning.

Branching Scenario Podcast with Mark Parry

Branching Scenario Podcast with Mark Parry

Mark Parry interviewed me for his podcast, Perspectives in Parryville. We talked about branching scenarios and Choose Your Own Adventure stories. His audience is primarily teachers, so this interview connects ideas from workplace training and education. This includes some discussion of the way feedback is used in branching scenarios, since feedback is an important topic for teachers as well.

Redefine Instruction: Storytelling for L&D

Interview on Storytelling for L&D

Kicking off the 2nd season of the Redefine Instruction podcast, I had a great conversation about storytelling for L&D with Sandhya Lakhanpal. We discussed a range of options for using storytelling and scenario-based learning. We also talked about my journey from teaching to instructional design.

Podcasts on instructional design careers, freelancing, and consulting

Freelancing for IDs

Freelancing for IDs with the Southern Nazarene University ID Lounge Podcast

This episode, Part 1, focuses on the practicalities of starting a freelance career and the necessity of a strong digital presence.

• Understanding the Terminology: The difference between W2 contracting and 1099 freelancing.
• Making the Leap: How I transitioned to full-time freelancing
• Where the Work Is: For those starting out, subcontracting through e-learning agencies is a recommended path because the agency handles the sales and project scoping, allowing the designer to focus on development.
• The Power of Networking: Rather than traditional “sales,” I emphasize “stealth mentoring” and being a “good online neighbor”.
• The “Required” Portfolio: A digital portfolio is considered the minimum requirement to be taken seriously as a freelancer.
• Portfolio Maintenance: Keep a “rainy day list” of portfolio updates to work on during slow periods.

The Future isn’t Flashier: It’s Smarter

The Future isn’t Flashier

I spoke with Holly Owens on the EdUp L&D podcast again in July, 2025. Our conversation covered the nuances of scenario-based learning, common pitfalls in instructional design, and the evolving role of AI in the field. I shared insights on how to effectively implement branching scenarios and the importance of aligning practice with real-world skills. Our conversation wrapped up with my thoughts on the future of instructional design—which is where Holly’s title of “the future isn’t flashier” comes from.

How I Became an ID

How I Became an ID

In this podcast interview with Dr. Robin Sargent, I discuss how I transitioned from teaching to instructional design and provide tips for new IDs.

Freelancing and Consulting

Podcast Interview on Freelancing and Consulting

Kristin Anthony talked to me about moving into freelancing and consulting, positioning yourself in the market, finding clients, pricing strategies, and managing my business.

Instructional Design Certificate or Master’s Degree

Instructional Design Certificate or Master’s Degree

I spoke Luis Malbas and TLDC about whether it’s better for your career to get an instructional design certificate or a master’s degree. This is part of their Certificates and Certifications playlist (members only for the videos for the whole playlist). We also talked about portfolios, as your portfolio is often as important (maybe more so) than your formal credentials.

Podcasts on AI and instructional design

Artificial Intelligence and Instructional Design: Insights from Christy Tucker

Listen to Part 2 of my interview with SNU’s ID Lounge podcast

This part of our conversation focused on AI and instructional design, published in April 2026.

• Practical applications of AI
• Analysis and skill mapping
• The efficiency illusion
• AI as a thinking partner
• Navigating ethics and privacy

From Generic to Custom: AI-Generated Images for Learning Content

IDIODC Podcast: AI-Generated Images for Learning

We talked about the challenges with generic stock photos and how my workflow has shifted to custom AI-generated images. I shared how I’ve used my blog posts for experimentation and building my skills with AI images. We talked about how to get consistency in images across several tools (the strategies are different in chat tools like ChatGPT and Nano Banana versus specialized image generators like Brushless and Recraft). We also talked about costs and subscription plans as well as legal and ethical issues.

AI Right Now

AI Right Now: Mindset to Learn Podcast

In season 2 of her podcast, Joanie Musser hosted a series of interviews about AI and how it affects our field. My discussion with Joanie was focused on my experiences as a practitioner. We talked about the tools I actually use like Midjourney and Claude. I shared how these tools can be helpful as well as some of the challenges I’ve experienced.

Podcasts on other topics: engaging learners, cost estimates, instructional design

Estimating eLearning Development Costs

Cost of eLearning Development based on Learner “Seat Time”: What the Research Says

This is a direct link to the YouTube recording for the IDIODC show (Instructional Designers in Offices Drinking Coffee). I reviewed Bryan Chapman’s research and the ATD research on estimating development time and cost.

Designing Engaging Digital Learning

Designing Engaging Digital Learning

In this podcast with Keith Quinn, we discussed a range of topics related to online and digital learning. We started with blogging as a way to “show your work” and reflect on your practice. The conversation touched on a number of other topics, including providing information versus practicing skills and using tools that aren’t intended for learning.

Emotionally Engaging Content

Emotionally Engaging Content

I spoke with JW Marshall on the Voices of E-learning podcast. We talked broadly about trends in workplace training. One of the themes of our conversation was how emotionally engaging content can improve learning outcomes.

Your podcast?

If you host a podcast related to learning and development, instructional design, elearning, AI, or other related topics, contact me about scheduling an interview.

Originally published 12/6/22. Updated 5/2/24, 5/15/25, 5/21/26.

The post Podcast Interviews on Scenarios, ID Careers, AI, and More appeared first on Experiencing Elearning.

For each pair of images below, the left or first image (depending on your screen size and email program) is Nano Banana 2, and the right or second image is from GPT Image 2 (the newest model in ChatGPT). In general, I prefer the Nano Banana 2 images; they feel a bit less staged and less like stock photos to me.

If you’re viewing this on my site and not via email, you can click each image to expand it.

Wheelchairs

Wheelchairs have been a problem for AI-generated images in the past, partly because so many stock images show able-bodied actors using temporary wheelchairs incorrectly. I was overall pleased with the results for both a “sporty wheelchair” and an “electric wheelchair.” There are still some flaws though—check the spokes on the wheels and the foot rests.

Style: Editorial photo
Subject: A Puerto Rican man, mid 20s, in a sporty wheelchair
Action: Working at a computer
Setting; A modern open office in a large city, high rises visible through the windows, other employees in casual clothes working nearby
Light: Natural light from large windows

Style: Editorial photo
Person 1: A Black woman, mid 40s, average weight, in an electric wheelchair. Wearing A pink t-shirt and jeans
Person 2: A Korean woman, mid 40s, wearing a long gray t-shirt and black leggings
Pose: Talking to each other, gesturing with their hands, friendly conversation
Setting: A trail in a suburban park with a playground in the background
Light: dappled afternoon light, partial shade from trees

Forearm crutch

Forearm crutches are one of the objects I’ve never been able to generate accurately in Midjourney. Even a year ago, ChatGPT struggled to show them accurately while also fulfilling the rest of the prompt. It’s much closer now. We often default to wheelchairs to represent disabilities in images, but mobility aids like leg braces and forearm crutches should be shown too.

Style: Editorial photo
Subject: Native American man, mid 50s, graying hair in a long braid
Clothes: A red button down shirt and gray slacks
Details: Using a forearm crutch
Action: Presenting at a small team meeting, explaining and gesturing with one hand
Setting: A small modern office conference room, a projector screen with a column chart behind him
Light: overhead florescent lights

Hearing aids

When I wanted an illustration of a woman wearing a hearing aid in January 2025 for a blog post, it took me over a dozen variations and edits to get it right. I’ve also had issues in the past where the AI models insisted that anyone wearing a hearing aid must be older and have gray hair, no matter what age I prompted for. This time, I got two plausible behind-the-ear hearing aids on my first attempt.

Style: Editorial photo
Subject: White woman, mid 20s, hair pulled back into a messy bun
Clothes: A dark blue business casual blouse
Details: Wearing a behind-the-ear hearing aid 
Pose: Typing, working at a computer with two monitors
Setting: An office cubicle with gray walls
Shot: Medium shot, head and shoulders

Cochlear implants

I think the cochlear implants here aren’t quite accurate; I don’t think they’d meet the approval of someone really familiar with these devices, especially at higher resolution. But this is closer than I’ve been able to get in the past. Because the implant is black in the Nano Banana image, it’s hard to detect.

Style: Editorial photo
Subject: Indian woman, mid 30s, braided hair
Clothes: A red tunic
Details: Wearing a cochlear implant
Pose: Sitting behind a desk, working on paperwork, focused expression
Setting: A private office in a bank, dark wood desk
Shot: Close shot, focus on the side of the head

Prosthetic leg

I set up my prompt for this image to put the subject in shorts so the prosthetic leg is visible without any weird artifacts like hiking up someone’s pant leg to show the prosthesis. I was also curious how the models would handle the setting of the YMCA; both models included the logo.

Style: Editorial photo
Subject: White man, mid 40s with a prosthetic leg
Clothes: A black athletic shirt and gray athletic shorts
Pose: Playing basketball, passing the ball to a friend
Setting: A gymnasium in the YMCA
Lighting: Overhead lights
Shot: Wide shot

Upcoming webinar on inclusive images

Disability representation is one of the topics I’m going to address in my upcoming Training Mag Network webinar, “Better Prompts, Better Representation: Inclusive AI Images for Learning.” This free webinar will focus less on the tools for AI images and more how to explicitly prompt for images that are more inclusive of gender, race, disability, and more. AI image models can reinforce stereotypes if you don’t prompt to counter their natural tendencies. That makes it important to know how to prompt well and what problems to watch out for.

Register for my free webinar on May 20, 2026 at 12 PM PT/3 PM ET.

What are your challenges with disability representation?

What challenges have you found with representing disabilities using AI images? Comment on this post or reply to this email. I’d love to have some more specific examples to test to see if it’s possible to generate with the most current models. What are the inclusive images you wish you had for your training?

The post Disability Representation in AI Images appeared first on Experiencing Elearning.

Principles of Multimedia Learning

Mayer’s Principles of Multimedia Learning – Educational Technology

This article summarizes Mayer’s Principles of Multimedia Learning with a few tips for instructional design aligned with each principle.

Personalization Principle – Using a conversational style in multimedia content makes learning more effective than a formal tone.

• Use a relaxed tone with contractions.
• Speak directly to the audience using “I,” “you,” “we,” and similar words.

—Dr. Serhat Kurt

Style of language Formal Versus Conversational

This guide provides a summary of the personalization principle with a focus on the writing or speaking style. A polite, conversational style is more effective for learning in general (with some exceptions noted). I appreciate the examples in this guide so you can compare the difference between formal and conversational style.

Water use for generative AI

How Much Water Does AI Use? An Expert Analysis of the Real Footprint.

The water use for AI data centers isn’t as big of a problem as it’s often made out to be. Don’t let the exaggerated claims about water use distract you from larger issues (including energy use, which is a real environmental consideration). We can push for improved efficiency and make choices about when and how we use AI to reduce that energy use

For 11 weeks, I tracked all of my AI use. One hundred sessions. I counted the tokens processed and applied publicly available numbers on per-token energy and water intensity from Epoch AI and operator-reported data from Microsoft and Google. Anyone can run this math.

In those 11 weeks, I built an iOS app from scratch and wrote policy briefs on extreme heat for nonprofits I work with. I produced documentary pitch decks and drafted a 15,000-word climate fiction piece about the Colorado River collapse. I used AI every single day, often for hours at a time.

Total lifecycle water footprint of all that work: about five gallons. That accounts for everything: the water used to cool the data centers, the water consumed at power plants to generate the electricity, and the water embedded in manufacturing the hardware.

…

You may have read stories about how data centers use lots of water, and how these massive warehouses of computer servers are being built across the country to help with the expanding use of AI. Here is the part that I think gets lost in the discourse. All U.S. data centers combined—not just AI, all of them—account for roughly 0.3 percent of total national water withdrawals. Agriculture consumes approximately 80 percent of Colorado River water.

AI images and voices

Seven Prompts No AI Image Generator Can Get Right

(If the link above to LinkedIn doesn’t work, try this version instead.)

Really interesting research on the limits of AI image generation. Hands and text are both much better than a year ago, but multi line text (especially with numbers) fails because text isn’t generated sequentially. AI images approximate rather than counting, and all models fail with prime numbers. Reflections are also approximate; there’s no geometry behind them.

Vois – Professional AI Voice Studio

While Vois doesn’t have as many voices as some other platforms, it has several other advantages. It runs locally on your machine, so there’s no risk of content being used to train AI. You can tag your script for multiple speakers, making it easier to manage dialogue. You can also buy just the credits you need rather than paying a monthly or annual fee, and you only use credits when you publish (not for each iteration and typo fix).

Automating repetitive design tasks

7 Ways to Automate Repetitive Design Tasks with Affinity and Claude

I use Affinity as my primary tool for editing images. Affinity now can connect with Claude to automate repetitive tasks like renaming layers and prepping files. It looks like a great way to speed up some boring tasks so you have more time on the fun work. This is currently in beta and free, but will probably become a paid feature later. Still, if it saves time, it may be worth a paid upgrade.

---

Upcoming events

Crafting Choices That Challenge: Designing Scenarios for Authentic Practice. July 15 at 12PM Pacific / 3PM Eastern. Register for this free webinar hosted by Training Mag Network.

Scenario-Based eLearning Design Certificate: Strategies to Engage, Motivate, and Transfer Skills. 3 online sessions starting August 5. Save $150 with Early Bird Registration.

Creating High-Quality AI Media: Prompting, Scripts, and Consistency. October 8. Register for the ATD AI Intensive.

DevLearn

Branch Smarter, Not Harder: Naming Conventions, Variables, and Bottlenecks in Storyline

• November 3, 2026 at 9:20 AM
• CE01: Storyline & Rise Pro Lab

Cohesive by Design: Generate Consistent AI Icons, Illustrations & Characters

• November 4, 2026 at 10:00 AM
• BYOD (Bring Your Own Device)

The post Multimedia Learning, GenAI: ID Links 5/12/26 appeared first on Experiencing Elearning.

One of the AI tools I’ve been using more lately is Flora. Instead of a chat interface like many AI tools, Flora gives you a canvas with multiple nodes where you can see and organize multiple results. Flora includes most of the main image generation models, so you can switch between tools easily. For my post comparing how different image models handled the same prompt, I used Flora’s built-in feature to run a single prompt with multiple models at once. In this post, I’ll show how I used Flora’s batch feature to generate character images in multiple poses quickly. After experimenting with this, I think this process can really speed up character image creation for scenarios. You can generate multiple poses of a character with consistent aspect ratios and even transparent backgrounds much faster than if you prompt one at a time in other tools.

This isn’t a sponsored post; I’m sharing my own experiences so you can see what worked for me. I’ll share some of my mistakes and lessons learned as well.

How do you create a character reference image in Flora?

The first step for generating character image sets is always generating a reference image. If your reference image isn’t good,

In this case, I wanted to replicate the visual style from another image I’d previously generated for a presentation. I uploaded that image to Claude and asked it to describe the style so I could use it in an image prompt. Claude gave me much more detail than I wanted, but I picked several of those phrases to include in my prompt.

I used a structured prompt with labels. Nano Banana and ChatGPT work well with this kind of prompt. It’s simpler than JSON prompting, and you get some of the same benefits of repeatability without needing to be so precise in your syntax.

Style: Contemporary digital painting, visible brushstrokes with flat color blocking, realistic body proportions, acrylic painting blended with vector-style cel shading
Subject: A Latino man, mid 40s, black hair, well-trimmed short beard
Clothes: Navy blue button-down shirt, khakis, brown Oxford shoes, black glasses
Pose: Standing, looking at the camera, smiling, relaxed pose
Background: Solid white

Here’s my reference image using Nano Banana 2 after a few attempts. I tried the ChatGPT Image 2.0 model too, but that was much slower.

How do you batch generate poses from a CSV in Flora?

Next, I wrote a list of poses and saved them in a CSV file. I used 8 simple poses for my examples for this post, but Flora can handle up to 100 at once in a batch. Practically speaking, I probably wouldn’t do more than 20 images at a time. If something goes wrong and you need to adjust your prompt or settings, it’s easier to adjust with smaller numbers of images.

• Talking gesturing with hands
• Joyous laughing
• Listening thoughtful arms crossed
• Disappointed slight frown looking down
• Frustrated upset
• Confused head tilted slightly
• Turned to the side talking and gesturing with hands
• Turned to the side listening with arms crossed

I uploaded the CSV with the list above to a Batch node in Flora. Flora separates out each row into a separate prompt.

How do you combine the reference image and batch prompt list?

This is where Flora really speeds up the process of character image generation. Instead of prompting 8 different times, I connect the node with my character reference image to my batch of prompts. With the batch, I can run 8 generations with a single click.

Running the batch of 8 prompts gave me these variations on the original character image. Note that even though it’s a single node on the right, there are actually 8 variations all stacked in the same node. Click each image in the gallery below to view a larger version. (If you’re reading this in email, you may have to view it on my blog instead.)

How do you remove backgrounds from a batch of images?

While I often generate scenario images of characters directly in scenes with context, sometimes it’s easier to work with separate character images with transparent backgrounds and then manually place them against a background image later. Flora makes this easy; removing the background is one of the tool options on image nodes. Since all of my character variations are in a single node, I just have to click remove the background in all 8 images in one step.

How do you scale this image creation process?

I generated another reference image for a character named Maia and used the same batch prompts for the variations. This time I used a 3:4 aspect ratio instead of a 1:1 ratio. I don’t know why Nano Banana matches the aspect ratio just fine in Flora but ignores it in prompts in the regular interface. But I get consistent ratios in Flora.

I also generated a set of images for a character named Ji-hun. These were a little less consistent, but still usable. I’m not sure sure she suddenly has wavy hair when she’s frustrated, and I don’t know why the laughing image wasn’t a full body picture. These are good examples of why you probably don’t want to run too large of a batch at once, and sometimes you’ll have to manually adjust and re-run specific images.

What are mistakes when batch generating character images?

I tried a few things that didn’t quite work out and that I’d do differently next time. Hopefully you can learn from my mistakes!

The standing pose is really similar in all of these results because I didn’t include any specific body changes in my prompt list. I think with a better list of prompts I could get more variety in the results. The facial expressions are good, but the poses don’t vary enough.

I didn’t realize that Ji-hun’s reference image wasn’t on a solid white background; it had a paper texture instead. The flaw in the character reference image meant I got more variety in the backgrounds later. That’s an easy fix if I remove the backgrounds, but it’s a good reminder to be more careful with the reference image.

With Maia’s character image set, I tried putting the reference image on a transparent background. I’d hoped it would generate all of the subsequent images also on a transparent background, but it it just put them on a solid black background instead. Again, an easy fix with removing the backgrounds, but I created some extra work for myself in needing a step to put them all on a white background for this blog post.

Some of these characters have shadows around their feet, but not all of them. That’s easy enough to fix in an image editing program though. I could probably add “no shadows” to the prompts as well.

Reflections

Overall, I think this is a really promising workflow for generating images for scenarios. I often generate a “shot list” for branching scenarios as a table (sometimes using Claude to generate a list based on my Twine export), so I already have a table describing scenes. This would work for generating images in context for scenarios all in a batch like this.

If I wanted to generate multiple sets of characters in standard poses with transparent backgrounds, this would be significantly faster than generating them one-by-one. Even if I just start with a standard set of poses and then prompt individually for a few specific images, this would speed up my process.

While I still like Midjourney for generating blog cover images and my initial reference images, I’m finding myself using Flora more often. It’s a definite shift in process, and I feel like I still need to spend more time to get really efficient at it. But I’m getting better the more I practice and refine my processes and prompts.

More on AI image generation

Learn more about generating character images using other AI tools.

• Generating Consistent Characters in the Midjourney Web Interface
• Using the Midjourney Editor to Change Character Expressions
• Testing Freepik Spaces for Character Image Generation
• Nano Banana (Gemini) Character Image Experiments
• Character Consistency Q&A

---

Upcoming events

Crafting Choices That Challenge: Designing Scenarios for Authentic Practice. July 15 at 12PM Pacific / 3PM Eastern. Register for this free webinar hosted by Training Mag Network.

Scenario-Based eLearning Design Certificate: Strategies to Engage, Motivate, and Transfer Skills. 3 online sessions starting August 5. Save $150 with Early Bird Registration.

Creating High-Quality AI Media: Prompting, Scripts, and Consistency. October 8. Register for the ATD AI Intensive.

DevLearn

Branch Smarter, Not Harder: Naming Conventions, Variables, and Bottlenecks in Storyline

• November 3, 2026 at 9:20 AM
• CE01: Storyline & Rise Pro Lab

Cohesive by Design: Generate Consistent AI Icons, Illustrations & Characters

• November 4, 2026 at 10:00 AM
• BYOD (Bring Your Own Device)

The post Speed Up Character Image Creation with Flora appeared first on Experiencing Elearning.

In branching scenarios with simulated conversations, sometimes I see people build every single exchange as an interaction. With that approach, if you’re controlling a character, then every single sentence that comes out of that character’s mouth has to be a decision point. It can get pretty tedious, both for the user and for the ID building it.

Example without a cutscene

Here’s what that might look like with some dialogue excerpted from the middle of a larger scenario.

Decision point 1

Decision point 2

Decision point 3

With a choice for every line of dialogue, that creates 3 decision points, each with 3 choices. That’s 10 screens that have to be built.

Example with a cutscene

Compare the version above to this version using a cutscene. This time, all of the dialogue continues seamlessly. You see some back and forth between the two characters before having a single decision point at the end. Instead of 10 screens, this is only 4 to build (the conversation shown here plus the results of the 3 choices).

See the cutscene in action

You can see the cutscene in action on its own or as part of the complete branching scenario. This full cutscene includes some additional dialogue prior to the script excerpt shown above.

I wrote about the specifics of building a simulated phone conversation previously.

Advantages of cutscenes

Simplify the scenario

The biggest advantage of showing a conversation in a cutscene like this is reducing the total number of slides (from 10 to 4) and decision points (from 3 to 1). That reduces time building the other options, as well as the time writing plausible distractors for the additional decision points.

Reduce cognitive load

Trimming extraneous decision points helps reduce the cognitive load on users. With a cutscene, they can see the conversation and focus their efforts on the decisions that matter. They don’t need to put in cognitive effort on the less important decisions of every line of dialogue.

Meet critical objectives

The cutscene moves the conversation forward quickly to get to the next critical point. In this particular example, the objective is to propose a road mapping. That needs to be a decision point.

The other parts of the conversation aren’t as critical for the learning objective. Therefore, it’s not as important for them to be decision points. I can collapse that dialogue into a single, non-interactive conversation. We need the context, but not everything has to be a decision point.

More tips on streamlining branching scenarios

For more tips for making branching scenarios easier and more manageable, download my guide on Streamlining Branching Scenario Planning and Design. This is a free 25-page PDF guide on how to save time and manage the complexity of branching scenarios while still providing rich, valuable learning experiences.

Originally published 3/3/2020. Updated 4/23/2026.

---

Upcoming events

Crafting Choices That Challenge: Designing Scenarios for Authentic Practice. July 15 at 12PM Pacific / 3PM Eastern. Register for this free webinar hosted by Training Mag Network.

Scenario-Based eLearning Design Certificate: Strategies to Engage, Motivate, and Transfer Skills. 3 online sessions starting August 5. Save $150 with Early Bird Registration.

Creating High-Quality AI Media: Prompting, Scripts, and Consistency. October 8. Register for the ATD AI Intensive.

DevLearn

Branch Smarter, Not Harder: Naming Conventions, Variables, and Bottlenecks in Storyline

• November 3, 2026 at 9:20 AM
• CE01: Storyline & Rise Pro Lab

Cohesive by Design: Generate Consistent AI Icons, Illustrations & Characters

• November 4, 2026 at 10:00 AM
• BYOD (Bring Your Own Device)

The post Cutscenes in Branching Scenarios appeared first on Experiencing Elearning.