· A recap of Vispero's presentations and announcements at CSUN.
· A preview of what's coming next.
· A sneak peek at some really exciting developments.
All this and more – on FSCast Issue 268.
OLEG SHEVKUN
Hi everyone, and welcome to FSCast Episode 268, for March 2026.
The 41st CSUN Assistive Technology Conference has just concluded in California. And our Vispero team was there and we had quite a number of presentations. And today's episode will be a recap of those activities. But also, we'll be looking ahead, and we'll be giving you a sneak preview of some capabilities that will be coming to JAWS and Fusion later this year.
But before that, let me go through some of the latest developments that are happening with Vispero software.
So in March, we released updates to JAWS, ZoomText, and Fusion. Now, in all of those products, you will notice some visual refresh as well as the start of our rebranding campaign. For example, FSCompanion is now Vispero Companion. FS Support Tool is now Support Tool. Our icons and graphics have been updated. Our website is also in the process of updating. The idea is to bring all of our software products under the Vispero brand. As you can read on our website, this is work in progress and you may still notice the mention of Freedom Scientific in our software products, but you will be seeing more and more references to Vispero.
Now, talking about functionality. In JAWS, we're now introducing something new via our Early Adopter Program, or EAP for short. That is an experimental version of native UIA support, primarily in web browsers, such as Microsoft Edge or Google Chrome. If you turn on this support mechanism via the EAP dialogue, JAWS will rely on user interface automation accessibility provided by the browsers. This support is currently work in progress. Some things may not work as desired, so you can always turn this feature on or off via the EAP dialogue, and of course, as with all experimental features, we would love to hear your input. There is the button to send us input right there in the early adopter program dialogue.
In ZoomText, the March update brings OCR capabilities. You can easily recognize PDF documents, graphics, pictures, camera input, and specific screen areas. The OCR button is now on the toolbar in ZoomText. You can read more about the OCR functionality in ZoomText in our release notes for the March update.
We hope you enjoy this updated version of JAWS, ZoomText and Fusion. Here at Vispero, we are already working hard on the next product update due to be released in May.
Now, to our JAWS power tip of the month. Well, this time it's actually a Windows power tip, and it comes from our colleague Rene Ludwig.
If you regularly work with multiple sound cards, you know how you can easily switch JAWS to output its voice to a specific sound device. To do that, you just press INSERT+SPACE followed by V, which is voice or volume, followed by C, and that stands for sound card.
Then with the arrow keys, you can move up and down through the list of available sound cards. This is especially helpful when you have multiple sound devices and JAWS is speaking through a wrong one. Maybe you don't even have the speakers connected to that wrong device, so your computer is silent. INSERT+SPACE V followed by C, followed by the vertical arrow keys, will allow you to actually find the device that can be used by JAWS.
But what if you want a command that will allow you to quickly change the output device in Windows? Yes, this keyboard command does exist, and it's actually provided by the operating system, not by JAWS. To quickly change your sound output device in Windows, press Control Windows V. This will open the quick settings dialogue and will put your focus on the list of currently connected devices for sound output. By using the arrow keys, you can easily and quickly change the device used by Windows.
Also, if you tab through this dialogue, this will give you a quick and easy entry point to other sound related settings.
I'd like to thank our colleague Rene Ludwig for contributing this power tip.
CSUN 2026 has just concluded in Anaheim, California. To talk more about this and about Vispero's presentations and announcements that were made during the conference, we're joined again by Ryan Jones, Vispero Senior Vice President of Global Accessibility Strategy and Advocacy. Well, Ryan, welcome to FSCast, and it is so good to see you again in our virtual studio.
RYAN JONES:
Thank you, Oleg. This is one of my favorite shows to be on. So I'm glad you put up with me coming on every once in a while, and we always have a good chat. I'm looking forward to it.
OLEG:
You always come from CSUN energized. I would suppose you come tired. Are those two mutually exclusive?
RYAN:
Yeah. It's this weird thing where you're exhausted on one hand because you're going 15, 16 hours a day, but at the same time, I really enjoy CSUN. I love getting to meet in person with so many of our users, our dealers, our partners, just other people in the field. So yeah, it's a weird thing to be completely drained and exhausted, but also looking forward to the future because we can see a lot of things that are coming down the road in the short term and in the longer term, and so, it's exhilarating from that perspective.
OLEG:
On our path to the future, sometimes we stop to think, to process, to evaluate. I understand that there's quite a bit of reevaluation going on here at Vispero. How much of this are you able and willing to share with our users? How much was communicated about this at CSUN?
RYAN:
I think it's important to remember that we, as Vispero, collectively, and all of us individually, are keen to listen and learn from the community and hear feedback. That feedback results in something all the time. Sometimes you see tangible things that the feedback results in, and sometimes it's intangible, but we really appreciate the feedback from users and customers. So, there's a couple of things that we announced specifically at CSUN this year that were directly related to feedback that we've heard with the 2026 versions of JAWS, ZoomText and Fusion. The first one is: when we released JAWS and Fusion 2026, we introduced a new feature called AI Page Explorer. This was a feature where JAWS can look at the contents of the webpage that you're on, run it through AI, and then generate some summary information about the page that you're on. So it could give you an overview of how the page is visually laid out, or how the page is structured. It would give you an overview of the main content on the page. It would even give you some navigational tips on how to interact with the page. Then you could also ask follow-up questions, kind of like with Picture Smart AI. So if you wanted to dig in a little bit deeper and find out specifics about what's on that page, you could do it through the AI Page Explorer.
When we released 2026, that feature was available in the Professional Edition and the Home Edition of JAWS and Fusion, but it was not available in the Home Annual edition. The Home Annual Edition, for those who aren't aware, is an edition that's only available in the US and Canada, and it's geared for people who are paying out of pocket for JAWS, ZoomText or Fusion for personal use at home. So in other words, you're not getting it through work, you're not using it for work purposes, you're using it for your own personal use at home. It's a lower cost option for people to use. AI Page Explorer was not available in that edition. We heard lots of feedback that people absolutely wanted that feature, and we heard discontent, to be frank. So, we reevaluated and decided to reverse the decision. AI Page Explorer is going to be available now in the Home Annual edition of JAWS and Fusion as well. That will take place most likely in our mid-May update. So not too far out, Page Explorer will be available in all the additions now of JAWS and Fusion.
OLEG:
Another area where feedback was coming has to do with accounts feature. Could you please outline some of that?
RYAN:
Yeah. We did announce as well that we are making some adjustments to the account registration process. We started this in the US, so this doesn't apply to anyone outside of the US, but when you installed and ran version 2026, you needed to create an account with the product. This is something we've never done before. Our products have never known Oleg versus Ryan versus someone else. Yet we need the accounts, because it opens the doors to a lot of great functionality coming down the road.
In fact, there are things being worked on right now that require the product to know who you are. There are things we plan in the future about that. We need to be able to comply with certain AI rules and regulations and validation. So there's a lot of reasons that we need the products to be aware of who is using them.
But when we released this originally, you had to create the account within one or two hours of first running version 2026 of either JAWS, ZoomText or Fusion. If you didn't create the account, the product would kind of go into a mode where it would only work for 10 minutes to allow you to either sign in or create the account.
So we also announced at CSUN this year that we're removing the enforcement requiring you to create an account within one or two hours of when you first use the product. We're still working through the details, but we did hear very clearly from the community that it was not enough time to create this account. So we're going to be changing that. I don't know the exact timeline, the engineering and product teams are working on that, but that was another thing we're doing in response to feedback from customers. I will say that down the road, we will be introducing the account functionality to other geographic areas, again, because we really think this opens up a lot of opportunities for feature functionality in the products to be more personalized as we go down the road.
OLEG:
On the hardware side, I understand that the Focus VI Braille display was also shown at CSUN.
RYAN:
Absolutely. We did have the new Focus VI 40, which is the next generation of our 40 cell braille display. It's not a revolutionarily new design, but there's some definite nice updates with it. So I'm sure we'll be on to talk about that more as it gets released, but we did have the Focus VI 40 there. A lot of people got to put their hands on it, try it. We had it working with JAWS. We have it working, of course, with iOS and Android. We had a lot of great positive feedback on it, and it was exciting to have something new there to show from a Braille technology perspective.
OLEG:
In software product updates, OCR has come to ZoomText, and AI is coming to our products in a big way. Yet we are talking about this AI development pretty early. Why so? What are we trying to achieve here?
RYAN:
It's a good question. So we always have to balance when we talk about something new, So it's not too early. And sometimes we have to change direction on things. But it's also really important to get feedback on capabilities that are being built. Even though we've been talking about this, and we've been adding AI features in JAWS for two to three years now, there’s some new functionality provided by AI. That's really the ability to actually have JAWS, in this case, perform tasks on your behalf.
So not just to tell you what's on the screen, but actually to help you perform actions. That's more than just screen reading. That's a whole new way of interacting with the computer and assistive technology. We know that it requires a lot of feedback. It requires a lot of engagement with the user community, with advocacy groups, with enterprises, and it's not just about the technology. We can build the technology quickly, but how you interact with the technology and how you put safeguards in and all these things that are not things that we've necessarily had to deal with in the past are confronting us all, not just from us as Vispero, but us as blind and low vision individuals and users of technology. So we're starting this journey very early so that we can start involving people and getting feedback all along the way of creating some of this new AI that we're working on.
OLEG:
You are in the audience of our AI presentation at CSUN. In your view, how was that received?
RYAN:
I think it was very well received. The session was standing room only, quite a number of our sessions from Vispero were standing room only, but we heard this at the booth with people talking about it. There's a real interest in how does this help me? How does this make my life when using a computer with JAWS easier? How do I accomplish more? How do I be more productive? How do I do things that I could not do in the past? So we have to be able to answer those questions as a company, and I think we have a great opportunity to do things that we could never do before. And there's definitely a lot of interest around that.
OLEG:
So Ryan, your job has to do with global accessibility and advocacy. Lately, there has been a feeling in the blindness community that Vispero is sort of disconnected from that community. How would you reassure our listeners? What kind of signal would you be able to send that, "Yeah, we are connected, and we're listening, and we're here for you?"
RYAN:
It's easy for me or anyone at the company to say that we're listening and we're connected. It's another thing to act it and live it out. I think the proof is in the pudding, so to speak, if I could use that colloquialism, but you have to see us out there. You have to see us responding to feedback. I think hopefully people have seen that just with the things that I just talked about of changes that we announced at CSUN this year, that were based directly on a significant amount of feedback that Vispero received over the past few months, and that led to these changes. So I think it's one thing to say it, but we have to live it out. I think you'll see that over the next coming months if you're at different conferences or activities that we're at. Again, we always want the feedback. We need the feedback. It's the engine that drives us forward. So if you don't see us responding, then you need to let us know. I mean, obviously we can't necessarily respond to every single thing that everyone says, but we need to be listening to the overall themes, the overall direction that people are giving us, and respond.
So I think we've done that. I'm proud that we've made some decisions recently that have changed the direction of our products, and I look forward to us continuing to do that as we go forward.
OLEG:
By the way, SightCity is coming up in just a couple of months. Ryan, I do hope to catch up with you in Frankfurt.
RYAN:
Absolutely. It'll be great to see you in person there.
One of the new faces here at Vispero is Shaun Ryan, our director of product management. Shaun has quite a responsibility here because he coordinates our latest AI developments and initiatives. Shaun is joining us on FSCast right now to bring us up to date. Hello, Shaun, and welcome to FSCast.
SHAUN RYAN:
Thank you, Oleg. Happy to be here.
OLEG:
Now, you are new to the world of accessibility, but you're definitely not new to the world of AI.
SHAUN:
That's correct. I actually joined Vispero in early December. So a lot of the industry is new to me, having an opportunity to learn a ton, which is something I'm really excited about. But AI is something I've been working on for the last few years. I've had a chance to use AI and intelligent automation to improve supply chain, to improve product design. Now, I'm really excited about the opportunity to see what we can do and how we can bring AI into accessibility.
OLEG:
Many of us still think of LLM, large language models. That's where you ask a question and you get an answer, but AI technology is just so much bigger than that.
SHAUN:
It is. So traditional LLM interactions, what you're thinking of, is like a chat interface where you prompt the LLM, and it goes out and searches the internet or some sort of data set and it produces a response for you. But what we're really exploring today is what's commonly known as Agentic AI. This is where you can employ agents to actually act on your behalf, which is something we're really excited about.
We're introducing Agentic AI into our Fusion products. This agent uses accessibility information from JAWS along with screen capture and other contextual data to build a really solid understanding of what is happening on the user's computer. It can use this understanding to do some pretty cool stuff, like explain in whatever level of detail the user would like about what's happening on their computer. It can guide the user through certain tasks. Importantly, the agent can actually do things that the user asks the agent to do.
OLEG:
So to me, that's both really desirable, but also dangerous because I'm letting some AI work on my computer. What are the guardrails here, if any?
SHAUN:
Yeah, absolutely. So we treat safety and trust as a concurrent path with product development, making sure that we're being very deliberate about our approach and how do we make Agentic AI safe, and specifically safe for the blind and low vision community.
As we all know, sometimes agents may misinterpret what a user asked it to do, or it may start doing something that it wasn't asked to do. While rare, those scenarios may happen, and it's absolutely imperative that we have a way to stop the agent in its track. So we've introduced with the press of a single button, the agent stops dead and it tracks mid-question, mid-action, it doesn't matter.
We've also introduced something referred to as human in the loop. This is a way for us to introduce human judgment and decision making into an agentic flow. There are certain risks associated with tasks that an agent could complete, right? Sometimes, as we say, clicking a button is just clicking a button, and other times clicking a button might mean submitting your tax form. So the point there is that there's very, very differing degrees of risk associated with both of those button clicks. So the agent is aware of risky behavior versus innocuous behavior. When necessary can prompt the user for confirmation, it can narrate to the user what it's doing, or it can sometimes decline a request if that request is illegal or extremely unethical.
OLEG:
There's also another source of uncertainty, and that's a user. I may be talking to my AI agent. I can say, "Hey, help me to configure the printer or press the button that would do this or that." But my request is not precise enough. It's not containing all the information. There may be ambiguity in the request. Do we have a strategy of how the agent is going to deal with that type of ambiguity introduced most likely unknowingly by the user?
SHAUN:
That's a great question. First, the agent is pretty darn good at navigating ambiguity on its own. It can infer quite a bit. Again, it has access to accessibility tree information. It has access to screenshots. It has access to focused apps. It can build a solid understanding of what's happening on the computer, and it does a pretty darn good job of interpreting what a user intends to do. As you mentioned though, there are times where it doesn't, and when there is ambiguity in a prompt, the agent will ask for clarity before proceeding.
OLEG:
That's the Human-In-The-Loop tool you mentioned earlier.
SHAUN:
Correct.
OLEG:
So just talking about the system capabilities, I understand we're still in the early development stage, but we are able at this time already to activate, to move focus, to type, to click, to ask the user and to do many other things. How do you see that toolbox or that tool set growing?
SHAUN:
As you mentioned, we've defined a lot of primitives that the agent can take today. So it can focus, it can type, it can click buttons, it can navigate to X, Y, coordinates on the screen. Believe it or not, those fundamental primitives combined to make up just about everything you can do on a computer. There are some gaps, and we're looking to fill those quickly. For example, drag and drop is an area we're exploring. So there are different tool calls that we're looking at introducing, but we believe we're able to achieve quite a bit with what we have.
OLEG:
Another headache for a product manager is the compromise between different categories of users, making sure their interests are taken into account. There may be a beginner user who may ask some very simple question, or they may be an advanced user with advanced requirements. Or on the other hand, there may be a user who just needs some learning or handholding, and there may be a user who needs the agent to take him or her out of a tight spot. How do you see all those categories together? Is there a particular type of a user you're catering to in the first place?
SHAUN:
Our primary focus out the gate is really intended for the user who needs to get out of a tight spot, as you mentioned, or we call them getting unstuck. This is not intended to be an always on agent that sort of lives in the background and is just kind of ever present while you're navigating your computer. This is really intended as something that you explicitly invoke. It's there when you need it. It's not there when you don't. A good way to think about it is on demand, off demand.
We know there are situations in a digital experience where a blind and low vision user may encounter what we call breakdown moments. An example of a breakdown moment would be visual and coding only. This is a situation where meaning is being communicated visually via layout or iconography or imagery, color. Another might be something like disorientation where the user is literally asking themselves, "Where even am I?" So in those situations, what we'd like to do is introduce a tool that, rather than having to follow some sort of very complicated workaround or bring in a third party application or maybe call somebody over for help or worst case scenario task abandonment, we'd like to introduce a tool that empowers them to go ahead and get through those stuck moments.
The nice thing about agentic solutions is that they're appropriate for beginner, intermediate, and advanced users. A beginner user can lean on an agent for basic tasks like navigation or to guide through steps. Whereas an advanced user who is already very proficient with JAWS can call on the agent to do things more complicated like navigate complex tool hierarchies to find a particular setting within an application. So I think there's value for all users there.
OLEG:
To set something really straight, you're not replacing a screen reader, such as JAWS, with agents.
SHAUN:
Absolutely not. In fact, we are explicitly building this agent into Fusion. I believe that that enables a few things. One, it makes the tool more powerful. Traditional agents that you kind of see popping up everywhere, they largely rely only on screen capture to build an understanding. Because our agent is built into JAWS or Fusion, we actually can leverage accessibility information as well as screen capture, which makes our agent a lot more powerful. It builds a much stronger understanding of what is going on the user's screen. The only thing that changes is that when and if the user would like the agents help, they can invoke the agent and ask for it, but everything they know and love about JAWS today doesn't change.
OLEG:
So Shaun, you've just returned from CSUN, and that's your first accessibility conference as far as I understand, but you've been to other technology forums or conferences. What is special about your CSUN experience? How does that compare with what you've experienced before?
SHAUN:
What struck me when I was at CSUN was how candid and personal my conversations were with everybody I talked to. Not to deride any other conferences. I've been to many over the years and I've enjoyed them all, but there's really no peacocking taking place. There's not a lot of like, "Look how amazing this thing I'm building is." There is a genuine sense that people are coming together to share really cool things that they're doing in the industry. There's a sense of a genuine desire to help and to improve tools and to basically make the world a more accessible place, which I think it's noble. I think it was a really cool thing to be a part of.
OLEG:
So you've been experiencing the agent at CSUN, and you've been experiencing that during its development, and in the demos and so on. Have you actually seen the positive feedback, or how the agent helps to solve an issue? Can we move from speculating, well, that will happen to, yes, we're developing that and I've seen it happen already.
SHAUN:
That's a great question. So I guess as a little bit of a lean in, I will say that when we were presenting the agent at CSUN, there were a couple of times during the demos where I heard audible, "Oh wows," coming from the audience, which was so cool to see. It showed that we're onto something, and that people are genuinely excited about what we're creating. I can tell you after the demo, one of the most common questions asked was, "When can I get my hands on this?" The first time you really see the agent in action where a user asks it to perform a task and it demonstrates that it understands what's being asked. It executes a series of actions that do what the user asked it to do. Then on the back end, instead of just stopping, it frequently We'll clarify what it did and provide some additional information to the user.
The reason why I think this is important is because we don't want the agent to become a primary mode of interaction. I believe that people will remain productive and efficient when they're continuing just to do things on their computer the way that they always do them. Going back to this kind of unstuck use case here. But when the agent summarizes what it did and communicates back to the user what was the problem in the first place, it empowers the user to solve that problem the next time they come across it. So now they become more capable, and I think that's really cool.
OLEG:
Well, Shaun, thank you for being with us on FSCast. Yeah, by the way, when is that agent going to be available?
SHAUN:
Right now, we are targeting a beta for sometime in early to mid-summer, and we hope to have it released to market shortly after that.
OLEG:
Shaun Ryan, director of product management here at Vispero.
Now, it's time for the demos. Before I demo anything though, let me say a couple of things. Number one, what we're showing right now is a very early version of our agent. We currently do not even know the official name of the agent, and therefore I will be referring to that either as agent or as Fusion AI. I'm also not demoing much of the user interface. Again, because that's currently in development, and what I now have here in our Vispero Virtual Studio may be different from what hits the computer near you when we release. Having said this, the basic capabilities are there, and this is some really exciting stuff.
So I chose to prepare three demos showing three different use cases.
The first demo is about setting up a keyboard. I have this older Logitech keyboard here. If I just connect this to a computer, I'm not going to be able to use the F1 through F12 keys directly, because by default, they will bring up this keyboard's multimedia and connectivity functions.
On newer keyboards, there's a shortcut to switch those modes. But on this one, to switch the mode, I would need to use a piece of software from Logitech. That software is not quite accessible. So I'm going to bring up our Fusion AI.
I'm on the edit field. I'm going to press ENTER to activate forms mode. And I type: “Help me configure this keyboard so that the function keys perform their original functions”. Then, I press ENTER to send the prompt.
Let's see what happens.
AI AGENT:
Clicking on the keyboard device shown on the left side of the screen to access its settings and configuration options.
Clicking the settings button to access the keyboard settings where I can configure the function keys to perform their original functions.
Clicking the toggle button to enable the option to use F1, F2, et cetera keys as standard function keys, which will make the function keys perform their original functions.
Perfect. I can see from the screenshot that the function key setting has been successfully enabled. The toggle button for use F1, F2, et cetera keys as standard function keys is now turned on, shown in blue/purple on the right side of the screen. Configuration complete. Your keyboard has been successfully configured to use the function keys, F1F12 as standard function keys.
Here's what was done. Open Logi Options+, The Logitech keyboard configuration application.
Selected your K780 keyboard.
Found the connected keyboard device.
Navigated to settings.
Access the keyboard settings page.
enabled the function key option. Toggled on, use F1, F2. Et cetera keys as standard function keys.
What this means:
your F1 through F12 keys will now perform their standard functions. F1 for help. F2 for rename. F5 for refresh. Et cetera. Instead of the special media functions they may have been performing before. if you need to access the special media functions, volume, brightness, et cetera. You would typically hold the FN key while pressing the function key. The setting has been applied and saved automatically by Logi Options+.
Your keyboard is now configured as requested.
OLEG:
Okay. Well, let me check. I'm just going to press F1.
JAWS VOICE:
F1.
OLEG:
It says F1, and there's no help message here, but JAWS echoed this correctly. So action completed.
For our next demo, let me show you a form which I found on a certain website some time ago. There is a survey with 10 questions. Each question has a set of radio buttons, allowing you to choose your answer. However, there is no “Next” or “Previous” button to navigate between the questions. I could not find that button either with Virtual Cursor or by tabbing through the page. So I'm going to pull up our agent. This time, just for a change, let me activate the Dictate function and ask the question by voice.
“I'm on a webpage where I'm unable to navigate to the next question. Can you figure out why I cannot see that button? Click it for me and explain why it's not in virtual cursor or in the tab order.”
AI AGENT
Clicking the right arrow navigation button to proceed to the next question.
Perfect. The click was successful. You've navigated to question two of 10. The page now shows a new question. How does your organization allocate resources for AI governance with five different answer options.
Summary, why the button wasn't visible
The root cause. The navigation buttons, left and right arrows, are not properly exposed in the accessibility tree. Here's why:
Missing semantic HTML. The buttons are likely implemented as styled DIV or span elements with CSS styling and JavaScript click handlers rather than proper button elements.
No Aria attributes. They lack the necessary Aria attributes, such as roll equals button that would make them discoverable to assistive technology like JAWS.
Result. They don't appear in the accessibility tree at all. They're not keyboard focusable, so they don't appear in the tab order. JAWS can't find them in virtual cursor mode, because they're not semantically marked as interactive elements.
This is a common web accessibility issue. The buttons work visually and with mouse clicks, but they're invisible to screen reader users because the developer didn't implement them with proper accessibility semantics. The good news, I was able to click the button for you using coordinate based clicking, which works around this accessibility gap. However, ideally, the website developers should fix this by either using proper button HTML elements or adding role=button” and keyboard event handlers to the styled elements.
You are now on question to. Would you like me to help you navigate further through the assessment?
OLEG:
So mission accomplished.
Sometimes the issue we're facing is not about getting stuck, or not about some inaccessible application. The issue is simply that we need to complete an action quickly, and we just don't remember how to do that.
I am now in a blank Microsoft Word document, and I need to set this document so that the page orientation is landscape and the document has two columns. So to make it faster, I'm going to call up our agent, and I'm just going to speak my request.
“Set up this empty Microsoft Word document so that it is in landscape mode and in two columns.”
AI AGENT:
Clicking the layout tab to access page setup options for orientation and columns.
Clicking the orientation menu item to access the landscape option.
Clicking the landscape option to change the page orientation from portrait to landscape.
Clicking the columns menu item to access the column layout options.
Clicking the two option to change the document to a two column layout.
Excellent. The document has been successfully set up. Looking at the screenshot, I can confirm that.
Landscape mode: The page is now in landscape orientation. You can see the page is wider than it is tall, which is the landscape format.
Two columns: the document is now formatted with two columns. You can see in the layout tab that the column settings have been applied.
The Word document is now ready with both settings applied.
Landscape orientation. The page is now in landscape mode.
Two-column layout. The document is formatted with two columns.
The document is empty and ready for you to start typing. The text will automatically flow from the left column to the right column as you add content.
Is there anything else you'd like me to help you with for this document?
OLEG:
No, thank you. Mission accomplished.
So now after all those demos, it's high time to invite Elizabeth Whitaker to join us here in our virtual studio. Hello, Elizabeth. Nice to see you again, and great to see you after CSUN.
ELIZABETH WHITAKER:
Hi, Oleg. It's great to be back. And yeah, CSUN was really interesting as always, so we really enjoyed that opportunity to be there.
OLEG:
Now, you did some demoing of our AI agent as well. I looked at those demos, and they would be kind of hard to present in a podcast form, because one of them was based on an Excel spreadsheet. So instead of doing this live here, can you describe how the agent helped you on that spreadsheet?
ELIZABETH:
Absolutely. So one of the things that I wanted to show with that demo was how powerful this agent can be. When you're in a situation where you know there's some information you're not getting, but you don't quite know what it is. That spreadsheet was a workbook that contained 15 tabs. Each tab contained information about a specific project. So each tab was named whatever the project was. Each project was broken down into tasks. So each sheet in that workbook, each tab had all the information about the different tasks in those projects. Then each task was assigned to someone different. So essentially, what I had was a sheet that had 15 different projects with tasks in each project that were assigned to me. So there were 15 tasks there, and I needed to know which tasks were assigned to me in each project. Then I needed to be able to go over to the status column and mark that. The status column had a dropdown where you could select whether you had started the project, whether it was in progress or whether you completed it.
So the first thing that happened was, I got into the Excel spreadsheet, and I missed something very critical that JAWS told me, and I just kind of missed the information as it went by, probably pressed CONTROL. So I was navigating the spreadsheet just fine, but I navigated over and was trying to enter my status for one of those tasks, and I couldn't access the dropdown. I didn't know why. I had no idea. So I used the agent to show how you could ask it a question, it would solve the problem. What had happened was the sheet had opened in protected view, so we couldn't edit it.
So the agent was able to tell me that, and also walk through the process to click on the enable edit button. That was the first piece of it.
Then the second piece was that the dropdown options were presented as icons, rather than text. So the agent was able to describe those icons to me, what they meant, and select the one that I wanted. From there on out, I would be able to navigate through and select them myself, because now I know what each of those icons stands for.
OLEG:
That is pretty cool. Another demo, that's calendars or date pickers. A good calendar or a good date picker on the web should be pretty easy to navigate. In the best case, you have an edit field, and you can just put in the date and you're fine. Not all of those cases are easy. Was that one of your demos where you had the date picker that was kind of tricky to navigate?
ELIZABETH:
Actually, what we used for that was a scheduler where I was trying to schedule an appointment. You could navigate the calendar, the date picker, just fine, but what would happen is, when you selected the day, it would select the correct date, and then you would go to select a time, and it would deselect the date you had selected. So, no matter what you did, it would always go back to the default, which was whatever the current day is.
So I had the agent go select a date, and then I asked it what times are available on the afternoon of March 16th, and it was able to tell me. And then I could have it select that date.
OLEG:
Now, do you have a concern that this kind of development will put you all in the training department out of a job?
ELIZABETH:
No. I mean, we're always using AI tools and we are providing training on AI tools. I think it's important to understand that these are tools. These are additional tools in your toolbox. So you still need to have an understanding of what's going on on the screen. That's where, of course, this agent gave us that.
What happens when you get that information is that next time you go to perform a certain task, you're kind of clued into what may or may not be happening there. Or now that it's told you how to perform that task, you know how to do it. So it can be faster to go back later and perform that task. It teaches you what's going on on the screen. It tells you about your environment, but there are always still things that you're going to need and want to do on your own, but the agent is there to help you if you get stuck.
OLEG:
Speaking of training, what do we have from Vispero in April?
ELIZABETH:
Well, we have two webinars. One's going to be on April 2nd, and that is our AI webinar. We've been focusing specifically on certain tools. Now, we're going to come back to that. We wanted to kind of take a break from focusing on the specific tool and start focusing on scenarios. What are some scenarios where you can use AI? Where can it help you be more efficient? How can you combine different tools in your daily tasks to really get the most out of AI?
So that's what we're talking about in April. We're going to do a few webinars like that, and then we will get back to our specific apps. For example, the Claude app. We've had some requests for that and more tools like that, Perplexity and some others. But we really just want to focus on showing you how to use AI to get the most out of these tools and to just be more efficient day-to-day.
So that's going to take place on Thursday, April 2nd at noon Eastern. Then we're also going to have a software webinar to talk about comments and revisions in Microsoft Word. That's going to be on Thursday, April 16th at Noon Eastern. We've covered this previously, but it's been a few years and those dialogues have changed. So we want to go over that again and make sure that you have the most up-to-date information.
OLEG:
Just a reminder that for all training related stuff you write to training@vispero.com, and for FSCast related stuff, you write to fscast@vispero.com.
Now, speaking of that, you may have noticed some rebranding going on. How does that affect FSCast? Well, we started in 2006, and in December 2006, Jonathan Mosen produced the first episode of FSCast. He edited the podcast for over 10 years, and then Larry and John Gassman took the baton. That was followed by Glen Gordon, and I had the privilege of taking the baton after Glen Gordon passed it on to me, and it's been FSCast.
Well, you're now listening to the last issue that bears this name. So remember it well.
What is that new name going to be? We're still thinking about it, but whatever happens at the end of April, we will be here with a new podcast from Vispero.
So Elizabeth, it's nice to be closing this historic episode of FSCast together with you, and thank you for bringing us up to date on what happened at CSUN and what's happening in training. I hope to see you next month.
ELIZABETH:
Absolutely. Thank you for having me, and I'll be back next month to talk to you about training again.
OLEG:
We hope to see you all next month. In the meantime, have a great time and all the best to you from Vispero. Bye.
ELIZABETH:
Bye.