Hands On AI Training

06/01/2026

Anthropic released Claude add-ins for Microsoft Word, Excel, PowerPoint, and Outlook this spring. They are now generally available, which means for most organizations on Microsoft 365, AI assistance is already built into the apps your staff uses every day.

Here is why that matters:

The most common pattern in early AI adoption is starting with enthusiasm and then quietly stopping. Not because the tools do not work. Because they live somewhere else. A separate browser tab. A different website. Copying your draft in, copying the result back. For people with full calendars and no extra time, that friction is enough to make the habit not stick.

That barrier is gone now for Office users. You can highlight a section in a Word document and ask Claude to strengthen the argument. You can describe a chart you need in Excel and have it drafted. You can get a first version of a slide in PowerPoint without leaving PowerPoint. The assistance is where the work is.

The skills that make this useful are still worth learning. Knowing how to give the right context and push back when the draft misses the mark still makes a real difference in what you get back. But the first step just became a lot smaller.

If you have been waiting for AI to feel like a natural part of your work tools, that moment is here.

handsonaitraining.com

05/29/2026

Agents are fantastic at working with data, but they are only as reliable as the workspace you put them in.

Think of it like onboarding a new employee. If you give them access to a shared folder that has last year's policy draft, this year's current version, a backup copy, and a version someone exported for a meeting, they have to figure out which one is real. Sometimes they get it right. Sometimes they don't.

Agents face exactly the same problem. For instance, if a folder holds three versions of the same document and one is outdated, an agent asked to pull information for a new project might draw from all three. The output looks fine until someone notices it includes details from a direction the organization moved away from two years ago.

This is fixable. As agents take on more real work, the folders they operate in need to be organized for agent use, not just human use. Most workspaces today were not built with that in mind.

I have developed a hands-on training that teaches teams how to optimize a workspace folder specifically for agent use, establish which files are the current source of truth, and run a simple maintenance process to keep that clarity over time. The approach works regardless of which agent tool you use, whether that is Cowork, Codex, or something else entirely.

The organizations getting this right now will be in a much stronger position as agent use grows.

Visit handsonaitraining.com to learn more or reach out directly.

05/26/2026

There is a survey out showing that 82% of small businesses have invested in AI tools. The typical small business is now running five of them.

And most aren't getting much out of it.

It is easy to see how it happens. You sign up for ChatGPT. Then try Copilot because it came with Office. Then add something for social media. Then something a vendor bundled in. Then one more thing someone mentioned at a conference. Before long you have five subscriptions, five learning curves, and nothing running at full speed.

More tools don't fix a skills problem. They multiply it.

I built Hands On AI Training because I watched capable, motivated people get left behind by exactly this pattern. Not because they weren't smart or didn't care. Because nobody showed them how to actually use this stuff.

One AI platform, used with real skill, does more than five you sort of know.

That is what the training is for.

handsonaitraining.com

05/25/2026

All gave some...some gave all. We honor the fallen today. Never forget!

05/22/2026

Two programs launched this spring that are worth watching if you work in local government, run a small business, or work at a nonprofit.

The National Science Foundation is building AI-Ready America hubs in all 50 states — state-level infrastructure designed to bring practical AI training to the general public, small businesses, and community organizations. Not a Silicon Valley program. Something designed for communities like ours. These hubs are still in early development and are not something you can sign up for today.

Google's AI Works for America is further along, offering free AI skills training to small business employees, state workers, and local residents through partnerships with public institutions.

I am not going to tell you these resources are available right now, because they are not fully there yet. What I can tell you is that when publicly-funded AI training does show up in your backyard, the organizations already working on their AI skills will get the most out of it. They will arrive knowing what they need, with staff who can actually apply what they learn rather than hearing it all for the first time.

The window where building AI skills requires real initiative is closing. It just has not closed yet. Getting started now still matters.

That is what I help people do.

handsonaitraining.com

05/20/2026

The federal government just opened $25 million in grants for AI workforce training, and I want to give you an honest read on it.

The Commerce Department's AI Upskill Accelerator Pilot Program is real money — $1 million to $8 million per award. But it is a competitive federal grant, and only 5 to 8 organizations nationally will receive funding. The requirements are real: you need an established coalition of regional employers, a 40% cost match, and the ability to demonstrate that AI is already reshaping jobs in your sector. This is not a quick application that a small nonprofit or company puts together over a weekend.

So why post about it? Because it is the first of what I expect will be more programs like it. Federal and state funders are beginning to put money behind AI workforce development, and that trend is going to grow. Organizations that are already investing in AI literacy and staff training will have a meaningful advantage when the next opportunity opens — and the one after that.

The organizations I train are exactly the ones these programs are designed to serve: nonprofits, public agencies, small businesses. Getting your staff comfortable with AI tools now is not just a productivity decision. It is how you become the kind of organization that can make a credible case for that funding when it matters.

I don't write grants. But I help organizations build the AI foundation that makes the grant case worth making.

handsonaitraining.com

05/18/2026

A recent report has a number in it I keep thinking about.

92% of nonprofits are using AI in some capacity. Only 7% say it is making a meaningful difference in how their organization operates.

The researchers called it an "efficiency plateau." I would call it a training problem.

The report found that most nonprofits using AI are using it the same way: one person, one task, whenever they remember to try it. Not embedded in how the team works. Not connected to the workflows where the real time gets spent. Just scattered individual use that never adds up to anything at the organizational level.

Meanwhile, the nonprofits that have moved AI into actual team workflows are saving several hours of staff time per week. At typical nonprofit wages, that translates to thousands of dollars a year in capacity pointed back at mission work instead of administrative overhead.

The technology is not the issue. Every organization in this study had access to the same tools. The difference is whether anyone sat down and figured out where AI actually fits, trained the staff to use it there, and built enough structure around it to make the habit stick.

That is the work. It is also where the 7% lives.

handsonaitraining.com

05/15/2026

Here is the habit that separates people who get real value from AI from people who are still frustrated by it.

When AI gives you a draft, do not just decide whether you like it or not. Actually work with it.

Most people land in one of two places. They accept whatever came back and move on, including the parts that are generic or slightly off. Or they feel like it missed the mark and end up doing the whole thing themselves, wondering why they bothered. Neither is the right move.

The better approach is to treat it like work from a smart new hire who does not know your organization yet. You would not publish whatever they handed you. You would read it, mark what needs to change, and tell them specifically why. "This is too generic. We serve rural communities in southeastern Minnesota, not metro areas. Can you try this again?" That is a revision, not a do-over.

AI will give you a solid starting point on almost anything. What it cannot give you is the judgment. The local context. The sense of what your audience will actually respond to. That part is yours. Your job is not to prompt and accept. It is to prompt, read, push back, and improve.

That second step is where you go from "this is kind of useful" to "this is actually saving me time."

handsonaitraining.com

05/12/2026

"Detroit’s panic makes sense when you keep in mind that BYD ships a new car every 18 months. GM and Ford still need three to five years."

The birth of a new General Motors car still begins with a sketch. But after that, the Detroit automaker is leaning into AI to supercharge development.

05/11/2026

Recently, ChatGPT got added to Excel and Google Sheets for everyone, not just enterprise accounts. Claude landed in Word, PowerPoint, and Excel. Same thing is happening on the Google Workspace side with Gemini.

I bring this up because the number one reason people give me for not making AI a regular habit is that it feels like extra work to get to. You have to stop what you are doing, open a different tab, explain your whole situation from scratch, and then paste something back into whatever you were actually working on. By the time you do all that, you have already written the thing yourself.

That friction is largely gone now for the tools most of us use every day. If you live in spreadsheets or Word docs — running budgets, writing reports, tracking program data, drafting grant narratives — the AI is now sitting right there. You highlight something, click a button, and ask your question.

What does not come pre-installed is the skill of knowing what to ask. That is still the actual work. But at least now you can practice it without leaving your spreadsheet.

If you have been curious about AI but have not made it stick yet, this is a reasonable moment to try again.

handsonaitraining.com