More Americans Are Turning to AI for Tax Filing Help

Tax season remains a routine few people enjoy. Even with tools like TurboTax, filing returns still takes time and carries risk. This year, more users looked to AI tools like ChatGPT to manage part of the process.

According to SesnorTower data, in April, 16% of ChatGPT activity involved tax-related questions. That figure rose four points from the same period last year. The increase suggests more users are willing to bring AI into personal financial tasks.

TurboTax users showed a stronger connection. About 36% of them also used ChatGPT in April. In 2024, that number was 13%. Engagement also ran higher. These users averaged more sessions and spent more time on the platform than others.

• Also read: AI Now Influencing Promotions, Raises, and Firings Across U.S. Workplaces With Minimal Oversight

Some may have used ChatGPT to simplify the filing process. Others may have looked for quicker answers or help with confusing terms. The shift points to growing interest in AI for detailed administrative work.

For software companies, the trend suggests a chance to adapt. TurboTax and similar platforms could use AI to support users directly within their services.

Use of AI for financial tasks is becoming more common. If growth continues, next year’s tax season may bring even higher numbers.

Note: This post was edited/created using GenAI tools. 

Read next: Digital Platforms Where Users Are Confused by Text Abbreviations the Most
by Irfan Ahmad via Digital Information World

Digital Platforms Where Users Are Confused by Text Abbreviations the Most

If you scratch your head when seeing EMBM (early morning business meeting), you’re not alone. Texting has completely changed how we communicate. We’ve gone from using full sentences to adding emojis, dropping punctuation, and replacing common phrases with an alphabet soup of acronyms.

Although we’ve incorporated some popular text acronyms into our everyday language, like LOL or YOLO, other abbreviations have become a bit more far-fetched, like 4AO, which means “for adults only.” Sure, text abbreviations might help save our thumbs, but some of these shortcuts have gone from simple phrases to unrecognizable clumps of letters that come at the cost of understanding.

Preply, an online language learning app that connects experienced tutors to learners worldwide, conducted a study to see which abbreviations are most confusing in Reddit threads, TikTok comments, and other digital media. So if you routinely think confusing text abbreviations are NAGI (not a good idea), read on to learn about the ones most Americans hate.

Platforms with the Most Complicated Acronyms

If you hop on Reddit, you’ll end up finding the most complicated acronyms, as 1 in 4 users are unsure about what those acronyms mean. Phrases like TLDR and NSFW are quite common on this platform, but not every Redditor will understand every acronym that gets used. Then when you dive into subreddits, the confusion can be so unclear that some users even go to Google to figure out what those text abbreviations mean.

X, also known as Twitter, houses the second most confusing text abbreviations, which isn’t too surprising since you’re limited to 280 characters per post. To fit thoughts into one post, you might employ a flurry of complicated acronyms, but doing so leaves 1 in 5 users saying they’re routinely confused. When you’re dealing with character limits, some people give up on clarity to avoid running out of space.

Aside from Reddit and X, the top 5 digital platforms with the most complicated acronyms are TikTok, online forums in general, and Instagram. Using acronyms certainly saves time and characters, especially if you face a limit on how much you can post, but it can take a toll on those you’re trying to reach. Communicating quickly online with text abbreviations isn’t always the most effective when it comes to making connections. It can leave users confused and frustrated, and if they don’t understand your content, they might hit that unfollow button.

Text Acronyms Americans Hate

Whether you use text abbreviations or not, some acronyms get used so often that they become a regular feature across platforms and even generations. But most Americans can agree that some abbreviations are so annoying, they refuse to even use them.

The most annoying acronym for Americans is LMAO, or laughing my a** off. While 31% of Americans say they find LMAO annoying, it’s also the third most common abbreviation to be understood. The second most annoying abbreviation is GRATZ, or congratulations, while the third most annoying abbreviation is IANAL, or I am not a lawyer. Plenty of people might do a double-take when faced with a text featuring IANAL. So, if you tend to use text abbreviations, you’re better off being selective about the ones you include.

While some of these abbreviations might be annoying, others are downright confusing. The text acronym Americans understand the least is EMBM, or early morning business meeting, with only 3.79% of Americans saying they understand it. The top 5 most confusing abbreviations also include AYOR (at your own risk), RLRT (real-life retweet), WTPA (where the party at?), and KPC (keeping parents clueless).

Interestingly enough, three other text abbreviations that Americans also understand the least are acronyms about evading the attention or understanding of parents. These include PITR (parent in the room), POMS (parent over my shoulder), and PAW (parents are watching). These three may be functioning exactly as intended.

Some text abbreviations are generally recognized, like LOL. Other abbreviations are far too long and confusing for most people to understand. Before you use an abbreviation, it might be best to think about who you’re communicating with to see if they will be able to understand your meaning.

Generational Abbreviation Trends

As the ones who grew up surrounded by technology like smartphones and tablets, you might expect Gen Z to be the generation that understands text acronyms the most. But it was millennials in this study who picked up on what each acronym meant the best. Growing up in the times of Y2K, it makes sense for millennials to be the most savvy about text abbreviations.

While Gen Z may not understand text acronyms the best, they certainly understand when to use those abbreviations and when to avoid them. Around 53% of Gen Zers surveyed had answered outright "no" to allowing text abbreviations in a work setting. The last thing employers would like to see in professional emails is the use of LOL, and Gen Z realizes that the use of text acronyms may not be a good idea “WH”(during work hours).

Those in older generations tend to have more of a problem with the use of text abbreviations. Around 23% of Gen Xers find text acronyms annoying and were the most annoyed by the use of LOL. According to the survey, 0% of Baby Boomers said they feel "very confident" in understanding Internet slang or abbreviations. What you might see as a way to streamline communication might be an entirely new language for someone else.

Text abbreviations can certainly be a fun timesaver to use. Not only do they allow you to text faster and use fewer characters, but they can also help punctuate emotions. While digital language may have evolved along with technology, it’s important to remember that not every person has changed along with it.

If the person you’re texting doesn’t understand the acronyms you’re using, you might waste time by trying to explain it to them. Some people could struggle with RBTL (reading between the lines) when it comes to these confusing clumps of letters, and you don’t want people to have to Google what you mean. So, the next time you’re communicating with someone online or in text, remember some language might not translate the way you’ve intended.

Read next: New Research Shows Language Choice Alone Can Guide AI Output Toward Eastern or Western Cultural Outlooks

by Irfan Ahmad via Digital Information World

Jack Dorsey Builds Offline Messaging App That Uses Bluetooth Instead of the Internet

Jack Dorsey has spent years talking about decentralization. Now he’s testing it in a hands-on way. Over the weekend, the Block CEO and Twitter co-founder pushed out a messaging app called Bitchat. According to him, the platform doesn’t use the internet, doesn’t ask for your number, and doesn’t store anything on a server.

The app works through Bluetooth, bouncing messages from one device to another. You don’t need a data plan or even a Wi-Fi signal. If someone else nearby has the app open, messages pass along through them. That chain can stretch across multiple people, giving the network a longer reach than regular Bluetooth. In good conditions, it can hit around 300 meters.

You won’t find usernames, chat backups, or inboxes from three weeks ago. Conversations are short-lived. Everything gets encrypted and stays local, then disappears. That’s the whole idea, skip the servers, skip the tracking.

It’s not the first app to take that route, as Bridgefy tried something similar years back. Protesters in Hong Kong used it to communicate when mobile networks were unreliable or cut off. Bitchat runs on a related principle, but it's built from scratch and follows Dorsey’s own design preferences.

The beta launched on Apple’s TestFlight, and the initial batch of testers filled up fast. Ten thousand slots went live, and then they were gone. Dorsey also posted a white paper online, calling the whole thing a personal experiment. The project explores Bluetooth-based networking along with message relays, encryption methods, and data transfer without any central control.

• Also read: Which Apps Top the List of Platforms Americans Want to Abandon?

The app already supports small group chats. You can create rooms, tag them with hashtags, and add password protection if you want to keep things limited. Down the line, Dorsey and his team plans to add WiFi Direct support. That would mean faster delivery and longer range, all while keeping things offline.

Compared to WhatsApp or Messenger, Bitchat barely resembles what most people think of as a messaging app. There are no profiles, no syncing, no cloud. It’s just one phone talking to another, quietly and directly.

Whether it catches on or stays a side project, Bitchat fits right in with Dorsey’s push toward simpler, user-controlled tech. No announcements, no fanfare, just a new tool passed hand to hand.

Note: This post was edited/created using GenAI tools.

Read next: Chatbots Are Pointing Millions Of Users to the Wrong Sites and Scammers Are Cashing In
by Irfan Ahmad via Digital Information World

Americans Most Eager to Ditch Instagram, Facebook, and X, Study Finds

Instagram may be a social media giant, but more people in the U.S. seem ready to walk away from it than any other platform. A new analysis of Google search trends shows that users are searching in droves for ways to delete or deactivate their Instagram accounts, putting it at the top of the list of platforms Americans most want to leave behind.

Between April 2024 and March 2025, there were more than 545,000 average monthly searches nationwide for terms related to quitting Instagram. That includes phrases like “delete Instagram account,” “how to deactivate Instagram,” and “Instagram addiction.” Facebook came in a close second, with just over 538,000 monthly searches, while other platforms like X (formerly Twitter), Snapchat, and Telegram trailed far behind.

The findings come from a study conducted by health and recovery publisher Recovered, which analyzed search behavior across all 50 states using Google Keyword Planner. The goal was to figure out not just which apps people were tired of, but also where in the country users were most likely to look for an exit from social media altogether.

To level the playing field between high- and low-population states, the researchers adjusted the data to reflect search volume per 100,000 residents. That’s where Hawaii stood out the most.

Hawaii Tops the List of Social Media Quitters

On a per-person basis, Hawaii led the nation, with an average of 568 monthly searches per 100,000 people related to deleting or stepping away from social platforms. For a state with a relatively small population, that’s a striking level of interest in disconnecting.

The most searched phrases in Hawaii centered around Instagram and Facebook, which seemed to dominate the list across the board.

California followed close behind, with 528 monthly searches per 100,000 residents, amounting to more than 205,000 searches. Which shows the desire to slow down or disconnect was clearly gaining traction in California.

• Also read: Creative Tech Roles Offer Higher Starting Salaries for 2025 Graduates

Nevada, Georgia, and a tie between New York and Texas rounded out the top five. All showed more than 500 monthly searches per 100,000 residents, with keyword trends echoing those in other states: users wanted to delete Instagram, shut down Facebook, or at least figure out how to take a break.

States like Colorado, Arizona, Washington, Virginia, and Oregon made up the rest of the top ten, each showing significant interest in going offline or stepping back from social networks. In every case, Instagram and Facebook appeared as the most frequently mentioned platforms in user search queries.

At the other end of the spectrum, Mississippi had the lowest interest in quitting social media, with just 320 monthly searches per 100,000 people.

State	Avg. Monthly Searches	Avg. Monthly Searches Per 100K People
Hawaii	8,145	568
California	205,598	528
Nevada	16,810	526
Georgia	56,610	513
New York	99,118	506
Texas	154,388	506
Colorado	29,598	504
Arizona	37,221	501
Washington	39,048	500
Virginia	42,916	492
Oregon	20,124	475
Illinois	59,404	473
New Mexico	9,953	471
Maryland	28,792	466
North Carolina	50,037	462
Utah	15,531	454
Delaware	4,676	453
Rhode Island	4,886	446
Tennessee	31,616	444
Massachusetts	30,865	441
Connecticut	15,613	432
Florida	96,981	429
New Jersey	39,608	426
Michigan	41,783	416
Alaska	3,052	416
Pennsylvania	53,768	415
Kentucky	18,774	415
Indiana	28,141	410
Vermont	2,642	408
Oklahoma	16,528	408
Maine	5,643	404
Ohio	47,494	403
Louisiana	18,422	403
Wyoming	2,340	401
South Carolina	21,468	400
Missouri	24,286	392
West Virginia	6,928	391
Arkansas	12,007	391
Nebraska	7,666	387
New Hampshire	5,388	384
Kansas	11,259	383
Minnesota	21,966	383
North Dakota	2,998	382
Idaho	7,448	379
Montana	4,273	377
Alabama	18,524	363
Iowa	11,344	354
South Dakota	3,240	352
Wisconsin	20,214	342
Mississippi	9,403	320

Instagram Fatigue Is Real and Facebook Isn’t Far Behind

While Instagram led the pack, Facebook wasn’t far behind in terms of user fatigue. Each month, hundreds of thousands of Americans looked up ways to delete or escape both platforms. Even X, Snapchat, Telegram, and TikTok made the list, but with far fewer searches.

Other apps, including LinkedIn, YouTube, Reddit, Messenger, Threads, Pinterest, and WhatsApp, generated fewer than 30,000 monthly searches each. That doesn’t mean they’re immune to user fatigue, but the urge to quit just isn’t as strong.

Rank	Social Media Platform	Average Number of Monthly Searches
1	Instagram	545,449
2	Facebook	538,533
3	X	155,862
4	Snapchat	88,234
5	Telegram	67,148
6	TikTok	28,526
7	LinkedIn	27,663
8	YouTube	17,384
9	Reddit	17,234
10	Messenger	16,909
11	Threads	8,300
12	Pinterest	7,195
13	WhatsApp	6,096

A Sign of Digital Burnout?

Be it social overload, algorithmic fatigue, or endless notifications, users in the United States seem increasingly ready to disconnect. The fact that so many are turning to Google for help shows that the urge to unplug isn’t just a passing thought, it’s something people are actively trying to figure out.

For now, Instagram and Facebook still dominate the social landscape. But if search trends are any clue, plenty of users are already halfway out the door.

James Wittams-Smith, founder of Recovered, explains that: 

“For those who engage with social media apps regularly, the process of scrolling and taking in images, posting and receiving positive affirmations from others, and other stimuli create the chemical, dopamine in the brain. The more you engage with social media and receive the rewarding dopamine hit it creates, the more your brain will seek it out to get another hit. This can lead to social media dependence and addiction.  

• Related: Which Tech Companies Make the Most Money per Employee?

"For those considering a break from social media, start small by setting aside a certain amount of time dedicated to social media per day. Using a timer or app that blocks social media can hold you accountable for this. Taking up new hobbies that are not technology-related is also a great way to take time off social media, as well as intentionally seeing friends and family in person. 

“Keep in mind these platforms are designed to get and hold your attention, so work on taking control of your usage instead of letting these sites control you. This way, social media can be something that enhances your quality of life, instead of diminishing it.” 

Notes: This post was edited/created using GenAI tools.

Read next: Chatbots Are Pointing Millions Of Users to the Wrong Sites and Scammers Are Cashing In

by Irfan Ahmad via Digital Information World

OpenAI Experiments with Study Feature and New ChatGPT Integrations

OpenAI is developing new features for ChatGPT that expand its use cases in both education and workplace research. Two recent updates in testing include a potential study mode and additional integrations for its research assistant tool.

One of the unreleased features, labeled Study Together, was first spotted in May and appears to be surfacing more often in recent builds. While it isn’t currently functional, references suggest it could allow students to prepare for exams either with friends or by using ChatGPT as a study companion.

Although OpenAI hasn’t shared details or a timeline, the mode signals a shift toward more collaborative or educational use cases within the platform. If released, it may offer an option for students to engage with academic material directly inside ChatGPT.

Alongside this, OpenAI is testing updates to Deep Research, an AI feature that automates the process of gathering and organizing information based on a user’s input. One of the most significant additions involves Slack integration, as spotted by Tibor Blaho.

The Slack connector would give ChatGPT access to messages within a user's workspace, allowing it to factor those conversations into research tasks. This could be useful for internal summaries, content drafts, or reports that draw from existing team discussions.

• Also read: AI’s Hidden Energy Bill: What a Single ChatGPT Prompt Really Costs

It’s not clear whether OpenAI has a direct agreement with Slack or if the feature is built using publicly available APIs, which come with known limitations. The Slack integration has not been widely rolled out and remains in the test phase.

Deep Research is also gaining support for other platforms. Recent builds reference an integration with Canva, and signs point to early testing of a Gmail connector as well. If enabled, the Gmail feature could allow ChatGPT to search inbox content or generate email summaries without manual input from the user.

These experimental additions suggest that OpenAI is continuing to position ChatGPT as more than a conversational tool. As the company works toward launching GPT-5, these smaller feature tests may hint at what future workflows could look like across education and workplace settings.

Note: This post was edited/created using GenAI tools.

Read next: Creative Tech Roles Offer Higher Starting Salaries for 2025 Graduates
by Irfan Ahmad via Digital Information World

Creative Tech Roles Offer Higher Starting Salaries for 2025 Graduates

A fresh look at early-career salaries reveals how quickly creative work is shifting. For graduates entering the job market in 2025, the best-paying creative roles aren’t necessarily in the arts or media, they’re in tech.

Topping the chart is a relatively new title: AI prompt engineer. The job pays around $70,000 to start, and even interns in the field average about $20 an hour. What’s driving demand is simple. These engineers help large language models perform better by crafting the questions that guide their responses. It’s part writing, part training, and very much in demand across industries that are racing to adopt AI tools.

Second on the list is UX design, where entry-level roles bring in about $65,000 per year. Interns can expect up to $30 an hour. These roles sit at the center of how people interact with websites and apps, and companies know that smart design keeps users coming back.

One step below that, social-savvy grads are finding opportunity in marketing. A TikTok consultant, for example, starts at $55,000. While intern pay is harder to pin down for that role, the full-time salary alone shows how valuable short-form content strategy has become, especially when brands need to connect with younger users who live on those platforms.

Traditional creative tracks haven’t disappeared. Copywriters start near $52,700, with internships paying around $17 an hour. In the gaming world, writing roles are pulling in $50,000 for newcomers, and internships can pay up to $25 hourly. These jobs tend to attract candidates who bring narrative talent along with a feel for interactive storytelling.

• Also read: Where the Biggest Telecom Investments Are Happening Around the World

A little further down the scale are roles like digital marketer, graphic designer, and video editor. Each offers a base salary around $45,000, depending on the company and region. The pay is decent, and internships for these positions range between $21 and $25 per hour. They’re also more widespread, which means more competition, but also more openings.

Social media content creators fall into the same income range, though the expectations vary. These roles often mix creative freedom with deadlines that move fast. At the bottom of the top ten list is journalism, which starts closer to $35,000. While the pay may be lower, the mission behind the job still pulls in graduates who care about facts, stories, and accountability. Interns here usually earn about $20 per hour.

What ties these jobs together isn’t just creativity. It’s the ability to pair ideas with tools. Graduates who know how to use design software, understand how digital platforms work, or can shape content for a specific audience are far more likely to land well-paying jobs early.

The research pulled from salary platforms like Glassdoor and LinkedIn, looked at average pay across entry-level roles and internships. Job Leads, the team behind the study compared dozens of job titles, including variations, to find what employers are offering now, not just what they promised in the past.

As more industries go digital, creative work is evolving with them. The best-paid roles go to those who aren’t just imaginative, but also strategic. If you’ve got an eye for design and a sense of how digital systems behave, you’re in a strong position, whether you're building AI prompts or branding a product on TikTok.

Job Title (Industry / Sector)	Intern Pay Per Hour ($)	Entry-Level Salary ($)	Mid-Level Salary ($)	Senior-Level Salary ($)	Executive-Level Salary ($)
UX Designer (Tech)	$30	$65,000	$95,000	$125,000	$170,000
TikTok Consultant (Marketing / Creative)	N/A	$55,000	$75,000	$95,000	$115,000
AI Art Prompt Engineer (Gaming / Media)	$20	$70,000	$95,000	$125,000	$150,000
Video Game Writer (Entertainment)	$25	$50,000	$70,000	$90,000	$110,000
Creative Director (Advertising)	N/A	N/A	N/A	$120,000	$160,000
Digital Marketer (Marketing / Creative)	$22	$45,000	$75,000	$149,000	$300,000
Graphic Designer (Design)	$25	$45,000	$69,000	$85,000	$125,000
Copywriter (Advertising / Marketing)	$17	$52,690	$78,000	$92,000	$139,000
Journalist (Media)	$20	$35,000	$55,000	$79,000	$95,000
Fashion Designer (Fashion)	$18	$33,000	$60,000	$68,000	$150,000
Video Editor (Media)	$21	$45,000	$55,000	$80,000	$120,000
Social Media Content Creator (Advertising / Marketing)	$18	$45,000	$60,000	$72,000	$85,000

Read next: Which Tech Companies Make the Most Money per Employee?
by Irfan Ahmad via Digital Information World

AI’s Hidden Energy Bill: What a Single ChatGPT Prompt Really Costs

From writing emails to answering questions and organizing data, generative AI has quickly become part of daily digital routines. But while these tools make work faster and more convenient, they also come with an unseen cost. Every time someone types a prompt into ChatGPT, energy is used to process the request, and with it comes a carbon footprint that’s easy to overlook.

Recent Surfshark research suggests that each ChatGPT query uses about 2 watt-hours of energy. That’s roughly the same power it takes to run a 10-watt LED bulb for 12 minutes or charge a smartphone with a 5-watt adapter for 24 minutes. Put another way, sending one query consumes as much energy as running a 1000-watt microwave for about seven seconds. Heating up a typical lunch takes three minutes, so you’d spend the same energy by making 26 chatbot queries.

The impact grows quickly when scaled. If every person in the United States sent just one query, the combined energy use would reach nearly 685 megawatt-hours. That’s enough to power 63 average homes for a full year, based on national residential electricity use.

Environmental concerns don’t stop there. Each prompt is estimated to produce 4.32 grams of carbon dioxide. That figure may seem small, but when multiplied by millions of users, the emissions become significant. A single day where every American interacts once with ChatGPT would generate roughly 1,479 metric tons of CO₂, about the same as what 322 gasoline cars emit in a year, or as much carbon as 1,500 people flying roundtrip between London and New York.

• Also read: Which Tech Companies Make the Most Money per Employee?

As global usage continues to grow, the pressure to optimize these models increases. By 2025, the number of generative AI users worldwide is projected to reach nearly 378 million, with 65 million new users added in just one year. That jump marks the fastest growth to date.

There’s still no single answer on exactly how much energy a chatbot query requires. Some studies suggest it could be as low as 0.3 watt-hours, particularly for newer, more efficient models. Others report figures closer to 3 watt-hours, especially with older or more complex systems. These differences reflect both advances in AI infrastructure and the challenges in measuring energy use directly, as most estimates rely on modeling, not public data from tech companies.

For context, Google Search uses about 0.3 watt-hours per query, making ChatGPT nearly seven times more power-hungry by comparison.

The carbon footprint depends on where the electricity comes from. Data centers still draw heavily from fossil fuels in many regions, which explains why a single chatbot query can carry a CO₂ price tag. Depending on local grid intensity, emissions per prompt can vary widely, from under a gram to over nine.

Researchers used standardized appliance ratings to create relatable comparisons. For example, they measured a 10-watt light, a 5-watt phone charger, a 100-watt TV, and a 1000-watt microwave, then calculated how long each could run using the same energy as one AI query. They also matched the U.S. population against national energy consumption figures to estimate how many homes could be powered if that same energy were used elsewhere.

None of this means people should stop using AI altogether. But as adoption accelerates, so does the need to make these systems cleaner and more efficient. Behind every AI-generated answer, there’s an environmental tab still being calculated.

Notes: This post was edited/created using GenAI tools. Image: DIW-Aigen.

Read next: Where the Biggest Telecom Investments Are Happening Around the World
by Irfan Ahmad via Digital Information World