Microsoft Power Automate Tutorial – How to get started

Power Automate (formerly Microsoft Flow) is one of the best automation tools if you are planning to find an IFTTT alternative for desktops. Here is how you can set up Power Automate on the web and desktop to get the most out of it.

Power Automate is an automation tool built into Windows 11 and accessible on the web as well as Windows 10. When Zapier, IFTTT, Automate.io, etc., work with online apps, Power Automate works with online and offline apps. In other words, you can perform various tasks that have no relation to the internet, such as creating a folder, renaming a file, copying content to another location, etc.

While accessing Power Automate, you would come across a term called flow, which denotes the type of template of your task. For instance, there are five different flows included in Power Automate, and they are Automated cloud flow, Instant cloud flow, Scheduled cloud flow, Desktop flow, and Business process flow.

There are mainly two parts to create a task in automation: trigger and action. Trigger lets you choose when to perform a task, and Action helps you decide what to do. All the tasks you will create or are already created come with these two things in the background. Once you are done with the task setup, Power Automate runs the task automatically in the background until or unless you stop it manually. 

In other words, it has the same workflow as IFTTT, Zapier, and Automate.io. All of them allow users to choose a trigger and an action accordingly. However, Power Automate has a handy option called On error. It lets you choose a different action when an error occurs in the trigger or action. All the other tools do not have this option for the users.

Power Automate has a wide range of compatible apps and services, allowing users to do more with an automation tool. Whether it is the web or desktop version, you could get the same set of options across both platforms. Talking about the specific apps, you can find Outlook, Excel, Word, etc. On the other hand, it offers different tasks based on variables, conditions, files, etc.

The best thing about Power Automate is that you can create a nested task. Let’s say you want to create a file in a certain location when you receive an email attachment, which is more than 10MB. If you cannot get it done in one task, you can create one task to fetch the attachment and another to create the folder.

If you are looking forward to an app with web and desktop support, Power Automate can be a good solution for you. Considering that, you can try using this service once to check if it fulfills all your demands or not.

To use Microsoft Power Automate on the web, follow these steps:

To learn more about these steps, continue reading.

At first, you have to visit the official website of Power Automate, which is flow.microsoft.com and sign in to your account. After that, click the Create button visible on the left side and select the Automated cloud flow option.

Here you can select other options according to your requirements:

The first thing you have to do is to name your task. Then, select a trigger according to your wish, and click the Create button.

Depending upon your trigger, you may have to sign in to the app or service. Following that, click the New step button.

Here you have to select an action for your trigger. For that, click on a task you want to perform in the given list.

As per your selection, you might have to sign in to the selected service again. Once done, you can click the Save button.

Now, you can run the task or flow in your account.

Before getting started, you should know that Power Automate Desktop works differently from the web version. In the desktop version, you have to create a nested task since there is no option to choose a trigger. In other words, you have to select one action as the primary action and the rest of them as the sub-action. The primary action works as a trigger, and the rest of them work as actions.

To use Power Automate in Windows 11, follow these steps:

Let’s check out these steps in detail.

At first, search for the Power Automate app in the Taskbar search box and open it from the search result. After opening, click on the New flow button.

Now you have to enter the name of your task and click the Create button. Once you do that, you can find another window on your screen with the task name in the title. Then, select the primary action from the left side.

You can either double-click on an action or use the drag-&-drop method to select it. Depending upon your selection, you have to fill-up the form on your screen and click the Save button.

Next, click the Subflows arrow and choose the New subflow option. You can enter the name of your subflow as well.

Then, you can select another action from the left side and click the Save button.

To explain the entire process, the first action will work as a trigger, and the Subflow will work as an action.

At last, you can click the Run button to test it.

Related: How to use Microsoft Power Automate Desktop to create a UI workflow.

Yes, Power Automate for desktop is free for Windows 11 and 10 users. If you cannot find it on your computer, you can download it from flow.microsoft.com. You have to use your Microsoft account followed by the installation.

Yes, Power Automate is included with Microsoft 365 subscription. If you have an active Microsoft 365 subscription, you can use Power Automate on the web for free or without additional charges.

Power Automate for web costs around $15/month/user. However, if you have Microsoft 365, you can use it for free. Also, Power Automate Desktop is available for free for Windows 11 and Windows 10 users.

Considering everything, Power Automate seems to be a handy automation tool made to make you more productive. Hope this tutorial helps you get started with this tool.

Read: Zapier is a free Automation tool and a IFTTT & Microsoft Flow alternative.

Meet the TikTok Girlies Helping ChatGPT Become Even Smarter

It’s not a stretch to say that every tech company and their mother launched chatbots powered by generative artificial intelligence this year. Take for example Google’s Bard, Anthropic’s Claude, Microsoft’s Copilot, and, of course, OpenAI’s ChatGPT. These chatbots are based on incredibly powerful large language models (LLM) and, in the short amount of time they’ve been accessible to the public, they’ve improved tremendously.

Such rapid refinement speaks to the power of LLMs and generative AI—but it’s also a testament to the behind-the-scenes work of hundreds of thousands of gig workers. These humans do a little-known job called data annotation, which entails rating and describing the inputs and outputs of AI models to help the chatbot “learn.” These tasks could involve reading and ranking AI-generated poetry, or labeling a menu’s text with what words correspond to food items, drinks, and prices.

Historically, companies have exported this labor overseas and paid workers pennies. But recently, they’ve been outsourcing data annotation work to a different and unexpected demographic—and it could lead to some unforeseen and even dangerous consequences.

First, there’s some Gen-Z slang you need to know: girlies are young, online, mainly female-identifying users—and Jackie Mitchell is definitely one of the girlies. As she chats with The Daily Beast on a video call, she sports a flawless deep red manicure that complements her auburn hair, with two loose strands tastefully framing her face. She’s wearing a sleeveless, cream-colored turtleneck and dangling pearl earrings—in other words, she looks like the opposite of a person who you’d assume trains AI chatbots.

Jackie Mitchell turned to data annotation to help supplement her income. Her TikTok videos talking about the gig has led to an influx of young women helping train powerful AI bots.

Courtesy of Jackie Mitchell

“I really did not ever anticipate doing anything like this, but I'm really into side hustles,” Mitchell told The Daily Beast.

When she was a college student, she wanted a bit of money to pay for date nights, but with school limiting her free time she didn’t have a lot of options for “traditional” jobs. So she turned to remote, on-demand work that she could do on her own time to bring in extra cash. She began taking short 10-minute surveys for money. One of these surveys turned out to be a data annotation project with a message at the end advertising work on a site called Data Annotation Tech.

That was two years ago. She’s since graduated and found a full-time job at a nonprofit—but still manages to work in data annotation as one of many side hustles. Lately, she’s been trying to make an extra $100 a day for 100 days to afford the down payment on her first house; since Mitchell works full-time, she’s doubled down on remote gig work to meet her daily benchmark.

In September, Mitchell started posting “day in the life” videos to TikTok describing her side hustles—including data annotation.

While she doesn’t mention the site she uses by name, data annotation makes an appearance in dozens of Mitchell’s videos in the series—including one post that has been viewed more than 4.4 million times. Dozens of users have commented on these videos, saying they’ve signed up for data annotation or asking questions about the work. Meanwhile, videos with the hashtag #dataannotation (which Mitchell does not use) have another 4.6 million views on TikTok.

Brin, another creator on the platform, said that the so-called “Jackie Mitchell effect” is indeed real: One of Mitchell’s videos popped up on Brin’s “For You” page and inspired her to take an unpaid exam through Data Annotation Tech.

“I watched it and I was like, hell yeah,” Brin, whose last name is omitted to protect her privacy, told The Daily Beast. “She's doing what she needs to do to make things happen—I want to do that. And I think I went into the exam five minutes after watching that video.”

Mitchell and Brin signed confidentiality agreements with Data Annotation Tech, so they can’t disclose any of the chatbots or projects they’ve essentially helped train. But they’re working with some of the biggest names in generative AI: Data Annotation Tech all appear to be owned by the same company called Surge AI, whose clients include OpenAI, Google, Microsoft, Meta, and Anthropic.

So, that means if you’ve used ChatGPT, Bard, or Bing in the past year to help you write your essays, answer questions, or even craft jokes, you’ve likely been chatting with a bot that’s been trained by a small army of TikTok girlies.

Mitchell said that she mainly completes projects where she writes, edits, or fact-checks a chatbot’s responses. She might have to compare two poems written by a chatbot to determine which of the two best answers a specific prompt, or take a response about the best seafood restaurants in Maryland and make sure the bot isn’t “hallucinating,” or fabricating parts of its response or making inaccurate statements.

There are other types of projects on the site, too, involving skills like computer coding and language translation. In theory, the bots can learn from human data annotation and craft responses that are more accurate, helpful, and safer.

In practice, data annotation can feel rote and monotonous, Brin said. “It's nothing incredibly fulfilling or anything really complicated for that matter. It's just trading your time for money—and sometimes you need that.”

In stark contrast to reports about data annotation workers who make between $1 and $3 per hour, both women said that the money—for them at least—has been pretty good. Mitchell’s rate has increased from when she began, from roughly $17.50 per hour two years ago to $20 to $30 per hour now. Brin said she has made about $1,000 in total since signing up for the site several months ago.

It’s no coincidence that they’re both young women either. According to Mitchell, women make up more than 80 percent of her audience on TikTok—a large portion of which are 18 to 35 years old.

“TikTok has made it really clear to me that my content is for the girlies, and I'm fine with that,” Mitchell said. “I'm a girl's girl in every sense. I’m doing this for the girls.”

The same goes for Brin. Her viewers are 95 percent female and “5 percent creepy men.” She speculated that data annotation work might appeal to single moms or young women with unpredictable schedules who are looking for extra income. Judging by the posts on social media, training AI chatbots is just another side hustle for this demographic.

But it’s potentially an unstable one. Experts told The Daily Beast that a single population taking over data annotation gig work could lead to lasting changes to AI models. There’s also the distinct possibility that this kind of work won’t be around for much longer—shifts in supply and demand could undercut workers’ rates, or even take humans out of the equation entirely.

The kind of data annotation that Mitchell and Brin do requires specialized skills in creative writing and editing, but not all forms of data annotation do. Some, like classifying objects as drivers, cyclists, or pedestrians to help train self-driving cars, might pay far less.

In June, The Verge and New York Magazine reported that an underclass of laborers, often in developing nations, are paid less than $5 per hour for data annotation work. A 2022 paper by three Google researchers interviewed dozens of data annotators and industry experts and concluded that the field suffered from “widely accepted yet questionable industry practices.”

The authors added that there exists “tension between the need for high quality data at low cost and the annotators’ aspiration for well-being, career perspective, and active participation in building the AI dream.”

Academic research, in contrast, often relies on small-scale, high-quality data annotation—and training a research assistant to annotate can be time-intensive and costly, according to Nick Pangakis, a political science PhD candidate at the University of Pennsylvania. Pangakis researches how LLMs—like the ones that power ChatGPT and other chatbots—can answer questions in the social sciences.

“There's a huge amount of training, and that's why it's so expensive,” he told The Daily Beast. And simply paying for data annotation does not guarantee its quality. “What's to prevent people from saying, ‘I’m getting $20 an hour, I'm just gonna click through it’? This is a major problem in survey research.”

The trend of young white women getting into data annotation as a side hustle might affect AI models in ways we’re not even aware of too. Training AI models is a new enough field that the research is still catching up to it. What is known, however, is that relying on a single demographic for trainers has the potential to skew a model's outputs—similar to how medical AI models trained on data from patients of one race might perform poorly when applied to all races.

“If it's just women on TikTok that are doing [data annotation], then there’s going to be a certain set of biases” they’ll introduce to the models, Pangakis said.

Given these very human problems with annotation, would it make sense to automate the job? Pangakis said that some in the field want to take humans out of the equation entirely but cautions against it, at least for now. Leaving annotation to the bots is cheaper and faster, but it leaves models vulnerable to reinforcing mistakes and bad habits.

“We don't necessarily know if it's performing well, and that's kind of dangerous, because you can have biased results and misleading conclusions without even knowing,” he said. Equally worrisome: remote, unsupervised data annotators could be using chatbots to complete tasks faster.

Data Annotation Tech’s code of conduct explicitly prohibits workers from using AI tools like ChatGPT to complete projects on the platforms. Still, without a way of detecting this behavior, an honor code is just that. There’s no way of actually stopping them from using AI.

Despite this, Mitchell said she’s been doing annotation work in good faith—though she doesn’t plan to continue much longer. Growing up amid various social media boom and bust cycles, she’s realized that it’s wise not to put all your eggs into one basket.

“I wouldn't count on data annotation being anything long-term, just like how any side gig I do is not long-term,” Mitchell said.

Even if some are picking up new side hustles, the TikTok girlies have left an indelible mark on AI chatbots. And so, who better to ask about the prospects of artificial general intelligence—a term used to describe AI models that can function as well or better than the human mind—than the girlies who spend hours each day engaging with bots?

Despite all of the justified and unjustified fears around the emerging technology, Mitchell, for her part, isn’t concerned that AI will surpass human creativity. None of the AI-generated responses, creative writing, or poetry she’s seen has even come close to the real deal. There will always be a place for people in the equation—whether it be in writing, creating art, or making a few extra bucks with your girls on TikTok.

“There’s no way to replicate the human experience,” she said.