Feed aggregator

It's February 14, 2025. You know what that means? Yes, Happy Valentine's Day, but there's also something else worth remembering today.

YouTube has officially turned 20 years old.

That's right, the internet's biggest video platform as well as one of the most trafficked websites in the world, was founded on Feb. 14, 2005. 

At least, that's when three former PayPal employees – Steve Chen, Chad Hurley, and Jawed Karim – got together and decided on a name: YouTube. As Karim later explained in an early re-telling of YouTube's history, "the work began" on YouTube on Feb. 14, 2005. The YouTube dot com domain name was registered 20 years ago on that date. 

However, the video platform you know and love today was certainly not what they had in mind. While the founders have previously shared other reasons behind starting YouTube, such as not being able to find clips of the infamous Justin Timberlake and Janet Jackson Super Bowl halftime show online, the reality is that the first iteration of YouTube was a dating website.

That's right, YouTube was first intended to be an online dating site.

YouTube in April 2005 as saved by the Internet Archive Credit: Internet Archive

In fact, you can go check it out and confirm it yourself by looking at the earliest archived version of YouTube that's saved in the Internet Archive's Wayback Machine from April 2005. Prominently featured at the top of the page, right below the login section, is an area to input your gender identity, the gender of the type of person you're looking to find, and an age range you're willing to date.

The YouTube dot com domain being registered on Valentine's Day is starting to make sense now, right?

Unfortunately, for Chen, Hurley, and Karim, no one signed up for YouTube, the dating website. The three even posted on Craigslist, looking for women to upload a video for $20, in hopes that would kickstart activity on the website, according to a Vice story from 2015. No one took them up on that offer either, according to the YouTube founders.

The first YouTube video ever

YouTube's co-founders decided to ditch the dating aspect of the site shortly after and opened up YouTube to all types of video uploads. With that pivot, the YouTube founders themselves decided to upload the very first video to the site themselves. On April 23, 2005, Karim uploaded "Me at the zoo," a video featuring the founder talking directly to a camera about elephants while standing in front of the animals at the San Diego Zoo.

YouTube would officially launch in December 2005.

Even though YouTube is now synonymous with online video, it actually wasn't the first video platform on the internet. Vimeo, for example, had launched a whole year prior to YouTube's launch.

However, the same week as YouTube's official launch to the public, Saturday Night Live aired its "Lazy Sunday" sketch by The Lonely Island. An unauthorized upload of the video was published on YouTube and went viral, cementing YouTube early on as the online destination for viral clips.

Less than one year later, in November of 2006, Google would acquire YouTube for $1.65 billion.

SAVE $30: As of Feb. 14, get Super Mario RPG for Nintendo Switch for $29.99, down from its price of $59.99 at Woot. That's a discount of 50%.

Opens in a new window Credit: Amazon Super Mario RPG $29.99 at Amazon
$59.99 Save $30.00 Get Deal

Need a little more whimsy in your life right now? We could all use some cheering up for one reason or another lately, and a great Nintendo Switch game starring everyone's favorite plumber is the perfect thing to do it.

As of Feb. 14, you can get a physical copy of Super Mario RPG for Nintendo Switch at Woot for $29.99. That's $30 off its normal price of $59.99, and a discount fo 50%.

Super Mario RPG is a remake of the SNES original Super Mario RPG: Legend of the Seven Stars. It's a much different adventure for Mario and the gang as it swaps platforming for role-playing as Mario works to track down the Smithy Gang. After crashing into the Mushroom Kingdom and stealing pieces of Star Road, someone has to bring the to justice, and that's going to be Mario, obviously.

Instead of running through level after level smashing question blocks and collecting coins, you'll work with a party of characters as you battle through the game and learn skills while fighting off colorful enemies. It's a challenging and addictive twist on the typical Mario formula, and a fantastic remake at that.

If you're ready to jump in with a fun alternative to Mario's greatest platforming hits, don't miss grabbing this remake. It's a surefire way to put a grin on your face if you're going through dark times.

OpenAI has removed the ChatGPT orange warning boxes that indicate whether a user may have violated its content policy.

Model behavior product manager Laurentia Romaniuk shared in a post on X that they "we got rid of 'warnings' (orange boxes sometimes appended to your prompts)."

Romaniuk also put the word out for "other cases of gratuitous / unexplainable denials [users have] come across," regarding ChatGPT's tendency to play it safe with content moderation.

Tweet may have been deleted

Joanne Jang, who leads model behavior, added to this request, asking "has chatgpt ever refused to give you what you want for no good reason? or reasons you disagree with?" This further addresses the issue that ChatGPT would previously stay away from controversial topics, but also flag chats that seemed innocuous, like one Redditor who said their chat was removed for including a swearword in their prompt.

Tweet may have been deleted

Earlier this week, OpenAI overhauled its Model Spec, which details its approach to how the model safely responds to users. Compared to the much shorter earlier version, the new Model Spec is a huge document, outlining its approach to current controversies like denying a request to share copyrighted content and allowing discussion that supports or criticizes politicians.

ChatGPT has been accused of censorship, with President Trump's "AI Czar" David Sacks saying in a 2023 All-In podcast episode that ChatGPT "was programmed to be woke."

However, both the previous and current Model Specs say, "OpenAI believes in intellectual freedom which includes the freedom to have, hear, and discuss ideas." Yet removing the warnings raised questions about whether this is related to an implicit change in ChatGPT's responses.

An OpenAI spokesperson said this is not a reflection of the updated Model Spec and the change doesn't impact the model responses. Instead, it was a decision to update how they communicate its content policies to users. Newer models like o3 are more capable of reasoning through a request and therefore hypothetically better at responding to controversial or sensitive topics instead of defaulting to refuse a query.

The spokesperson also said OpenAI will continue to show the warning sign in certain cases that violate its content policy.

SAVE 25%: As of Feb. 14, you can get up to 25% off some of their bestselling sleep essentials. Plus, get an additional 10% off with code WELCOME10.

Opens in a new window Credit: FluffCo Up to 25% off at FluffCo Get an extra 10% off with code WELCOME10 Shop Now

Investing in high-quality bedding is one of the easiest ways to get a better night’s sleep. A quick swap from cheap, scratchy sheets and a pancake pillow set to luxe, hotel-grade linens, and a dreamy, down-filled comforter pillow combo can make all the difference. The only problem? Upgrading your sleep setup can get expensive fast, especially if you're looking to bring home that five-star hotel feel.

Luckily, FluffCo's Presidents’ Day Sale is offering shoppers up to 25% off some of their bestselling sleep essentials, including the ever-popular Down Feather Pillow and Down Blended Comforter. You can also get an additional 10% off with the code WELCOME10. And, if you spend $150, you’ll score a free pair of hotel slippers. (Spending $200 will get you a free linen spray, and $300 scores a free robe.)

SEE ALSO: Sleepmaxxing is TikTok's latest wellness trend, but does it really work?

So, whether you're in the market for a whole new bedding set or want to treat yourself to an extra-cozy pillow, now’s the time to shop.

Here are some of our top picks from the FluffCo Presidents' Day Sale:

• Down Alternative Pillow — $69 $115 (save $46)

• Hotel Lounge Robe — $119 $200 (save $81)

• Flat & Fitted Sheet — $149 $499 (save $350)

• Down Alternative Comforter — $149 $325 (save $176)

SAVE $159.01: As of Feb. 14, get the Apple Watch Ultra 2 for $639.99 at Amazon. That's a 20% discount and its lowest price yet.

Opens in a new window Credit: Amazon Apple Watch Ultra 2 $639.99 at Amazon
$799.00 Save $159.01 Get Deal

If you've been trying to decide on a new smartwatch, now's the time to lock one in. The Apple Watch is a great all-rounder of a pick that works well for just about everyone. But if you're looking for something more rugged and feature-rich that can do just about anything, the Apple Watch Ultra 2 is your best bet, and it's on sale right now for its lowest price yet.

As of Feb. 14, get the Apple Watch Ultra 2 for $639.99 at Amazon, down from its usual price of $799. That's a discount of 20% and the lowest price we've seen so far. This price applies to the GPS + Cellular model with a Rugged Titanium case and Indigo Alpine Loop band.

The Apple Watch Ultra 2 is the perfect companion for active users, thanks to its ability to withstand a variety of conditions while you run, cycle, hike, dive, and get out there and see the world. It has a 3000-nit display and a rugged build that you can take anywhere without having to worry about the repercussions. Bad weather? No problem. It keeps on ticking.

You get a whole suite of advanced tracking features, sleep analysis, and much more in addition to health features like fall detection and ECG monitoring as well. And if you don't want to use it for those things, it's still great for using for making calls, texting, and using your favorite wrist-based apps. And with up to 72 hours of battery life on Low Power Mode? You've got plenty of time to use it.

Time to lock your Ultra 2 in if you want to get it for the cheapest we've seen.

SAVE $350 As of February 14, the Lenovo Yoga 7i is on sale for just $549.99 ($350 off the usual price). It's our favorite Windows laptop under $1,000, so snag this deal while you can.

Opens in a new window Credit: Lenovo Lenovo Yoga 7i (Intel Core Ultra 5 125U, Intel Graphics, 16GB RAM, 512GB SSD) $549.99 at Best Buy
$899.99 Save $350.00 Get Deal

Convertible laptops are the best of both worlds. Why buy a laptop and a tablet separately when you can have both at the same time? Buying one is a no-brainer, in our opinion.

And it just so turns out that one of our favorite convertible laptops is sitting at a deep discount right now — the Lenovo Yoga 7i is just $549.99, which is $350 off the original price. It's one of our favorite budget laptops in general, as well as our top pick for Windows laptops under $1,000, so we definitely recommend snagging this deal while you still can.

SEE ALSO: Best MacBooks 2025: Which model would we buy?

In our review, we highlighted the Yoga 7i's satisfying, responsive keyboard, port variety, and performance for the price. "If you want a convertible laptop that can double as a tablet, handle basic tasks with ease, last a decent while on a single charge, and you don’t mind a dim display or occasionally glitchy trackpad, the Yoga 7i 14 is a great deal at $899. If it’s on sale, it’s a fantastic deal," wrote Mashable Freelancer Sarah Chaney. While the hinges could be stronger and the display could be brighter, they definitely aren't dealbreakers, especially at this price. If you're currently in the market for a new laptop and want the most bang for your buck, the Lenovo Yoga 7i is a great option while it's on sale for just $549.99 ($350 off). Grab yours while the deal is still live.

It's Valentine's Day, and what's more romantic than Pornhub? (Don't answer that.)

The ubiquitous tube site sent Mashable a map of the top relative searches across the U.S. in the week leading up to Love Day. The term is that searched more often in each state than in others.

Credit: Pornhub

In New York, searches wanted more "emotional" videos, and Wyoming was wholesome with "hugging." Contrast that with BDSM in Oregon and strap-on in Georgia.

SEE ALSO: How to unblock Pornhub for free

You'll notice that many states are grayed out. Pornhub blocked 17 U.S. states (so far) due to age-verification laws, which are oftentimes requirements to submit ID to view websites with over a third of explicit content. The Supreme Court recently heard a case about age verification, but we'll have to wait a few months for the verdict.

For now, if Pornhub is blocked in your state, you won't know what terms make your fellow statespeople's hearts flutter, but you can use a VPN to access the site.

SAVE 20%: Get up to 20% off sitewide during Ruggable's Presidents' Day Sale. Use the code STARS.

Opens in a new window Credit: Ruggable 20% off sitewide with promo code STARS Ruggable's Presidents' Day Sale Get Deal

As the weekend gets closer, more and more retailers are gearing up for the first big shopping event of the year — Presidents' Day. While this federal holiday is supposed to be a time to reflect on the legacy of America's founding fathers, it’s also become synonymous with some of the biggest sales and discounts of the year (especially on mattresses and other home goods — you can check out our full list of the best Presidents' Day sales).

One sale we were pleasantly surprised to see this year is Ruggable's Presidents' Day sale, where shoppers can get up to 20% off sitewide (select rugs are 15% off) on all of the brand's popular washable and stain-resistant rugs. Just be sure to use the code STARS at checkout to take advantage of the savings.

SEE ALSO: Honest review of Ruggable: Are the washable rugs worth the price?

Shop large area rugs, doormats, runners, and more at a discounted price to refresh your floors just in time for spring.

Rug deals to Shop at Ruggable:

• Desert Sumac Rug (3'x5') — $103.20 $129 (save $25.80)

• Morris & Co. Strawberry Thief Indigo & Ivory Rug (3'x5') — $143.20 $179 (save $35.80)

• Justina Blakeney Josephine Midnight Rug (3'x5') — $143.20 $179 (save $35.80)

• Linear Striped Multicolor Rug (4'x6') — $151.20 $189 (save $37.80)

• goop Salerno Soft Black Re-Jute Rug (3'x5') — $186.15 $219 (save $32.85)

Check out the best Amazon deals of the day as of Feb. 14: OUR TOP PICK Apple Watch Ultra 2 (GPS + Cellular, 49mm) $639.99 (save $159.01) Get Deal BEST LAPTOP DEAL Samsung 14-inch Galaxy Book4 Pro $699.99 (save $650) Get Deal BEST BLUETOOTH TRACKER DEAL Tile Sticker (2-pack) $33.99 (save $11) Get Deal BEST SOUNDBAR DEAL Roku TV wireless soundbar $99.99 (save $50) Get Deal BEST BEAUTY DEAL Shark FlexStyle HD430 $209.99 (save $90 with on-page coupon) Get Deal

If you love discount shopping, we've got just the thing for you this Valentine's Day. We've rounded up five of the best deals we could find at Amazon, with standouts from top brands like Apple and Samsung. While Valentine's Day isn't exactly known for its great sales, Presidents' Day is. And since it's right around the corner as well, consider these deals a gift.

Here are our top picks for the best Amazon deals of the day on Feb. 14. Be sure to take a peek at our picks from Feb. 12 and Feb. 13 as well — many of those are still live, including a $38 pair of Sony wireless headphones and 23% off the Apple Pencil Pro.

Our top pick: Apple Watch Ultra 2 Opens in a new window Credit: Apple Apple Watch Ultra 2 (GPS + Cellular, 49mm) $639.99 at Amazon
$799.00 Save $159.01 Get Deal

While it may not be the best choice for everyone, the Apple Watch Ultra 2 is better than the basic Series 9 (and the Series 10, for that matter) in every way. It features a more durable titanium case, a programmable action button on the side, built-in LTE, a depth gauge and water temperature sensor, an emergency siren, and twice the battery life at up to 36 hours (up to 72 in low power mode). If you're a hiker, runner, swimmer, or diver, it packs advanced features the other Apple Watches lack. Snag the 49mm version on sale for just $639.99 at Amazon as of Feb. 14. That's 20% off and its lowest price on record.

Samsung 14-inch Galaxy Book4 Pro Opens in a new window Credit: Samsung Samsung 14-inch Galaxy Book4 Pro (Intel Core Ultra 5, 16GB RAM, 512GB SSD) $699.99 at Amazon
$1,349.99 Save $650.00 Get Deal

If you're in the market for a new laptop, you could do a lot worse than the Samsung Galaxy Book4 Pro. It packs an Intel Core Ultra 5 processor with Intel Arc Graphics, a 3K AMOLED 2x touchscreen, AKG Quad speaker, a new Knox security chip, and up to 18 hours of battery life. The 14-inch model weighs only 2.71 pounds and packs 16GB of RAM and 512GB of storage — all for a grand total of $1,349.99. If you pick it up at Amazon as of Feb. 14, however, you'll pay just $699.99. That's 48% off and its best price to date. You'll also get Assassin's Creed for free.

Tile Sticker (2-pack) Opens in a new window Credit: Life360 Tile Stick (2-pack) $33.99 at Amazon
$44.99 Save $11.00 Get Deal

Tile's smallest Bluetooth trackers are best for finding household objects like your remote, glasses, keys, and other stuff that regularly goes missing. It sticks onto items (hence the name) to allow you to keep tabs on things around the house. Connect your device to the app and tap the tracking button to locate your lost items. Your Tile Sticker will sound off a loud ring when it's within Bluetooth range (up to 250 feet) or show you its most recent location on a map when it's outside that range. As of Feb. 14, you can grab a two-pack of these helpful trackers for only $33.99. That's 24% in savings and the lowest price on record.

Roku TV wireless soundbar Opens in a new window Credit: Roku Roku TV wireless soundbar $99.99 at Amazon
$149.99 Save $50.00 Get Deal

Pair this Roku soundbar with your Roku TV sans cables and "drastically improve the sound," as our friends at PCMag (also owned by Ziff Davis) put it in their review. While it's not hefty on extra features, it's inexpensive and can add big, mostly balanced sound to your cinematic setup. There are even simple sound settings that lower loud commercials, boost voices, and optimize night listening. Already pretty affordable at $149.99, it's on sale at Amazon for just $99.99 as of Feb. 14. That's 33% in savings.

Shark FlexStyle HD430 Opens in a new window Credit: Shark Shark FlexStyle HD430 $209.99 at Amazon
$299.99 Save $90.00 with on-page coupon Get Deal

The Shark FlexStyle is our favorite dupe of the ultra-expensive Dyson Airwrap. Mashable Shopping Reporter Bethany Allard found that it offered a nearly one-to-one experience with the Airwrap, noting that she couldn't imagine a reason why you'd choose Dyson's more expensive option instead. The HD430 system includes the auto-wrap curlers, a styling concentrator, an oval brush, and a paddle brush, making it ideal for straight and wavy hair. It's usually $299.99, but thanks to an on-page coupon, it's on sale for just $209.99. Just be sure to apply the coupon to get the discount.

None of these deals catching your eye? Check out Amazon's daily deals for even more savings.

If there’s a masked ball in a Regency romance, there’s always going to be mistaken identities and secret trysts aplenty. And that’s what’s in store for Bridgerton Season 4, which released a sneak peek today.

Netflix has released a glimpse of the coming season that's all about Benedict Bridgerton, shown as part of a Valentine's Day panel with showrunner Jess Brownell, stars Luke Thompson, Yerin Ha (Sophie Baek), and Golda Rosheuvel (Queen Charlotte).

In the clip, you'll spy all your favourites back in their costumes and elaborate wigs, but there's particular focus on a masked ball and the line "Don't worry, I will not tell my mother." Is it warm in here?

The first three seasons of Bridgerton are now streaming on Netflix. Season 4's release date is TBC.

Have you ever been frustrated by a friend or family member who never remembers things you've told them? That shouldn't happen with Google's Gemini chat bot anymore.

Google announced in a company blog post that Gemini can now recall past conversations you've had with it in order to enhance its responses in future ones.

It can summarize previous conversations or at least reference them when relevant to your current query. This should reduce the need to search through old chats you've had with Gemini, according to Google.

SEE ALSO: How to turn off Gemini in Gmail

You can also delete your chat history, if you'd prefer. Right now, this feature is only available to Gemini Advanced subscribers, and only in English.

Google says it will roll out to Google Workspace Business and Enterprise customers, as well as those who speak unspecified other languages, soon.

That's all fine, just as long as it doesn't use your previous conversations to fabricate facts about cheese.

On dating apps, everyone is trying to put their best face forward. For some, that's easy — maybe they're genetically gifted, have a knack for taking great selfies, or have friends who are photographers. But for many, filters and editing apps have become a go-to, even if they only use them for minor touch-ups. Now, with AI photo generators rising in popularity, I'm worried we're getting too far from reality.

At a dinner with friends last month, the conversation turned from AI to the fact that photos on dating apps don't always match a person's appearance. Sometimes, it's easy to tell, as they look different in every shot. Other times, it's much more challenging. That gave me an idea. What if I were to take an AI-generated headshot of myself and add it to my dating profile? After all, it's me — but also not me. Would people notice or call me out for it? 

The use of AI-generated photos on dating apps is not a new one, there’s an AI tool that creates photos specifically for dating apps, and people have started noticing AI-generated photos on other people’s profiles. Hinge's parent company Match Group declined an interview about whether it's aware of AI-generated photos being used on profiles and if it's doing anything about it. Instead, a representative said that Tinder, which is also owned by Match Group, is working on "an AI-powered tool to help select photos for profiles," which sounds similar to the one Bumble already uses.

To understand the appeal of using AI-generated photos in dating profiles, I decided to give it a try. And the result was soul-crushing. I, naively, never thought that the AI photo would be so popular compared to my other, real photos.

The rise of body dysmorphia

One of my favorite subreddits to peruse is r/InstagramReality. Redditors post edited or filtered photos and videos of other people — celebrities, influencers, and normal folks — they have found on social media, usually next to ones of what the person looks like in real life. 

Often, the editing or filtering is comically overdone and obvious. But, sometimes, Redditors have to point out where things warp or changes have been made to show that it’s not real.

It’s scary how rampant and excessive editing is getting. The comments in the subreddit often touch on body dysmorphic disorder (BDD) because, really, how else could people put some of these photos out there and think they look normal?

“It is actually a pretty common psychiatric disorder,” Evan Rieder, a board-certified psychiatrist and dermatologist in New York City, told Mashable of BDD.

About two to three percent of the population has BDD, said Rieder, and it is most likely under-diagnosed because people don’t always bring up their concerns to a mental health practitioner — and aestheticians don’t always ask why someone wants to change something. “They will basically fixate on something that is invisible or barely perceptible to the outside observer,” Rieder said. Mostly, this is the skin and hair; muscles are also significant for men. 

BDD isn’t just about poor body image, either — which is often influenced by societal ideals or because of how a person has been treated. It is a serious condition that is listed in the Diagnostic and Statistical Manual of Mental Disorders. At some point, an individual with BDD has performed a repetitive behavior such as mirror checking or reassurance seeking, according to the manual’s fifth and most recent edition, or a mental act such as comparing their appearance with that of others.

“In the ‘90s, the only person that was able to give filtered images was, say, the art director at Vogue magazine, and they would make everybody look perfect.

“Now you have these apps, and you can swipe away all your imperfections and give this hyper-idealized version of yourself that's there for public consumption,” said Rieder. “People often make themselves look unrealistic or cartoony. And they think that some of these things are possible.”

How I generated my AI photos

You may have already heard about Remini, an app that was popular earlier this year when its AI-generated photos of people's potential future children started popping up on Instagram and TikTok. People have been using it this past week to generate '90s-style yearbook photos, as the EPIK app that originated the trend has sometimes been overloaded.

I decided to use Remini for my experiment rather than an app like Lensa, as the AI photos it produces look more realistic (albeit edited). It also has a ton of options for different backgrounds and clothes. To start, I had to sign up for a 1-week free trial, pick my gender: female, male, or other, and upload at least eight photos of myself. I chose two pictures from the past few years, then took the rest over the next few days in various locations and outfits from different angles.

These are four of the photos I used to train the AI model and generate the AI headshots in Remini. Credit: Saira Mueller

Once the photos were uploaded, I chose my first “model image” — the style the AI bases its generated pictures on — and had to wait about 17 minutes for it to work (the first time you generate after uploading photos takes longer, after that it drops down to about three minutes each time).

SEE ALSO: Hinge reveals the top 25 successful prompts

Then, it was a case of repetition to get photos that were anatomically and physically correct — the AI is bad at things like hands and generating necklaces, sometimes completely changed my body shape and weight, and one time even put my arm through a wall — and that also didn’t alter my facial features too much. I spent an embarrassing amount of time generating these photos. Ultimately, I chose four I wanted to use on my dating profile.

The four AI-generated headshots from Remini that I picked to use on my Hinge profile. Credit: Saira Mueller

The experiment, I decided, would be done on the dating app Hinge, as commenting on people’s photos and written prompts is at its core, and users still do so regularly. I put the images on my profile for one week each, rotating them out every Monday, alongside five of my actual, non-AI generated photos to see how they would stack up — tallying every like and comment.

The reality (pun intended) of the situation

I uploaded the first photo with the blue turtleneck at 10 p.m. Within 14 hours, I had five likes — all for that photo. Two days later, 60 percent of my likes were for that photo (the remaining 40 being split between my other five shots). By the end of that first week, the AI image had 17 likes and two comments. My non-AI photos, together, got 14 likes and two comments.

These are the five non-AI photos from my Hinge dating profile. Credit: Saira Mueller

That trend continued for the next three weeks, with the AI photos consistently performing better than (or at least equal to) the other photos collectively. If anyone noticed something off about the AI photos (for example, my ears disappear, and the irises aren't spherical in any of them), they didn't comment on it.

Here are six of the comments from men on Hinge about some of my AI-generated headshots. Credit: Saira Mueller

It was, frankly, off-putting to see these generic photos outperform the ones that show the real me — that show my personality, my interests, and my life. Yes, my written prompts got a healthy dose of likes and comments throughout those weeks. Still, it’s hard to remove the physical aspect of a dating profile from that equation — even on an app that foregoes swiping.

The reason the AI photos were likely so well received was summed up perfectly by ChatGPT.

I asked ChatGPT what the four AI-generated headshots have in common. This was its answer. Credit: Saira Mueller

When I asked it to point out the two AI-generated images out of four photos from my dating profile, ChatGPT said that it can be "challenging to definitively ascertain which ones are AI-generated based purely on visual inspection" but that "some common indicators of AI-generated photos can include perfect symmetry or too-perfect features."

The struggle between seeing perfect features all around you and looking in the mirror and only noticing all of your perceived imperfections is something I've dealt with since I was young. Perhaps it was because I began reading women's magazines at 12. Maybe it's because I was bullied endlessly around the same time due to my skin color, height, and acne, thanks to puberty. It could also be due to my Asian mother, who commented on everything from my hips to my eyebrows for as long as I can remember. Part of it is probably due to social media.

There’s nothing worse for a perfectionist with diagnosed anxiety than having a running tally of all the ways you fall short of your ideal physical self. I asked my housemate about three of the perceived issues on my list, things I’ve struggled to accept for a long time, and she said she has never noticed them — and couldn't see the issues even at the moment when I was sitting in front of her telling her about them.

Looking at the same thing over and over has an effect on you. Rieder explained it as the Mere Exposure Effect. While repeatedly noticing your perceived imperfections in the mirror can have a negative impact, so can the social media you consume.

“The more you see something, the more you find it to be attractive,” said Rieder of the Mere Exposure Effect. “The more images I see of glass skin, the more I’m going to be desensitized to it, and the more I'm going to think it's less strange, and the more I'm going to find it attractive.” Glass skin, for those who are unfamiliar, is a Korean beauty trend in which an individual’s skin has the appearance of glass due to how even-toned, poreless, and luminous it looks.

“Imagine this sort of information overload, showing you what you want to see or what you're thinking about — and then you're clicking on an image, and [social media is] showing you more and more and more of that,” Rieder said. “That can actually warp what you find to be attractive. And we see that all the time in the aesthetic fields.”

The recent shift towards magazines, brands, and shows like Naked Attraction putting natural, unedited bodies out there is a good step. And while I don’t recommend putting AI-generated photos on social media, let alone dating apps (seriously, don’t do it), I can understand why it’s an appealing concept.

It’s taken me a long time to love myself for who I am — something I’m still working on. And while I often wish we lived in a world where editing apps and filters didn’t exist, it’s our reality, and we have to be more aware of how it affects us.

Now that the AI-generated photos are off my dating profile, I get excited every time I see a notification. I know that the person who is interested likes me for who I am — no filters, no editing, and no AI.

You know that feeling of anguish when you’re trying to think of the perfect opening line on a dating app? Or when a match replies but you’re stressing at work and simply don’t have the capacity to think of something witty-but-chill-but-hilarious? Well, there’s an app for that now.

Welcome to the world of AI dating assistant apps. No, this isn't about fake photos or virtual girlfriends. Instead, apps like Rizz and WingAI are generating opening lines and conversation replies – all you have to do is upload screenshots of the profiles or messages. You can select different conversation tones from "NSFW" to "casual", and the aim is to remove some of the burden of modern dating.

Feeling uneasy? Think this marks the end of authenticity in the sorry minefield that is already the online dating scene? That it’ll never take off? Well, actually, a 2024 report by dating app Flirtini shows that a whopping 55 percent of people have already used AI, such as ChatGPT, to generate messages to send to dates. A 2024 study by McAfee shows that over one in four Americans (30 percent of men, 27 percent of women) also already use AI to enhance their dating profile – the next venture for Rizz and WingAI, and a feature already available on YourMoveAI.

Why are people using AI dating assistant apps?

So, who’s actually using these apps then? Have you been reeled in by AI without knowing? And is it really as creepy as it seems? I speak to regular AI dating app users and other daters to find out whether it’s the next frontier of modern dating. 

Artem, a 35-year-old accountant from Miami has been on Rizz for around eight months and uses it to replace pestering the group chat. "My friends are much funnier than me, so I’d often workshop clever responses with them for a match," he says. "But obviously, I can’t lean on them all the time and the AI app is available 24/7."

SEE ALSO: Why some people on dating apps just want to be 'pen pals'

As for how it works, for memorable openers he uploads a screenshot of his match’s dating profile and chooses the tone of "Rizz" – a relatively spicy, cheeky mode. The app suggests a selection of opening lines relating to the profile and Artem hits refresh, or adds in key words to focus on, until he finds one that suits. For help with conversations already on the go, he uploads a screenshot of the exchange and receives suggestions for his reply. Artem rarely copies and pastes directly, instead he uses the suggestions to get his own creative juices flowing.

"The other day it gave me the opening line of, 'If you had a third nipple, where would it be?' which I obviously didn't use, but it got me thinking," he says. "The girl I was talking to had loads of tattoos, so I asked her, 'If you had a tattoo and you could never show anyone, what would it be?'" Another reportedly successful opener – this time, for a woman who worked for a non-profit – reads: "Hey, so I'm hosting this charity event next week for people who can't reach orgasm. If you can't cum, please let me know."

I’ll just let that sit for a bit. 

I really did try to keep an open mind, but both of those lines pretty much turned my stomach. The cringe factor is toe curling, obviously, but the sexual innuendo side is just downright uncomfortable – a consistent problem I noticed when messing about on the apps myself. Practically 99 percent of the suggestions I got were unsendable, and if I’d received them from someone else, it’d likely be an immediate block . (Notably: "Hey there, amazing mind! Your intellect is arousing, but so is that naughty tongue of yours. wink") The apps claim to be trained with the help of male and female dating coaches, but surely no woman would sign that off?

SEE ALSO: The best dating apps for introverts who prefer books to bars

Artem insists it pays off with practice, though. His main reason for loving the app is down to dating fatigue – basically burnout and exhaustion from meeting new people and it going nowhere. "There's an element of exhaustion, for sure,” he says. “Chatting with one person is cool and fun, but it can be emotionally draining when you end up speaking to five or eight people at once while you try to decide who to choose to focus on." Using Rizz means he can devote less bandwidth to coming up with the kind of clever one-liners that reap success in the Tinder world.

This echoes a wider sentiment from app daters these days – that the process has somewhat lost its sheen and is now just a means to an end. A 2024 study by Forbes Health found that 79 percent of Gen Z report dating app burnout, with women feeling it more at 80 percent compared to 74 percent of males.

"If I have one more, ‘Hey, what's up? How are you?’ conversation I'm done."

That’s why Taylor, a 30-year-old vet from Florida, uses WingAI. "If I have one more, ‘Hey, what's up? How are you?’ conversation I'm done," she says. "Using the app opens the conversation up more, in a fun way. It helps with the fatigue and makes it more interesting."

Doesn’t all of this just create more admin in the dating process, though? Apparently not. "I was already going back and forth on Google, searching for fun openers relating to certain jobs," she says, a method Artem also used to use. Taylor uses WingAI to inspire her first lines on Bumble but, like Artem, always edits the suggestion to make it sound more like her personality. "I don’t always like the ideas, but they give me an idea of where to go with it," she says. "I make sure it’s a bit of a hybrid between me and the app."

As for the authenticity of it all, Taylor thinks it’s only as inauthentic as the rest of online dating. "Dating on the apps is always fake and superficial to begin with. You match based on what someone looks like and the highly curated persona they put out there," she says. "As long as AI isn’t having the entire conversation for you, I don’t see the problem."

It's a fair point. Polishing your profile can take hours, weeks and more friends than you’ve had dates. The witty Hinge prompt that sold you on someone could easily be their bezzie mate’s humour, not theirs. There are the outright liars to consider, too: According to a recent Forbes Health survey, close to one in five adults polled claim to lie about their age, with men even more likely to tweak their numbers at 23 percent compared to 19 percentof women. Other reported fibs include income (14 percent), height (12 percent), as well as hobbies and interests (14 percent). Think of all the people paying for premium dating app accounts as well, which offer various leg-ups like sorting your potential matches by who’s most likely to be responsive – whether that’s a new member or someone active today.

"If I completely relied on the app suggestions and didn’t work in my personality then, yes, it would be an inauthentic representation of myself."

Bangalore-based Dhruv, 25, agrees with Taylor. "If I completely relied on the app suggestions and didn’t work in my personality then, yes, it would be an inauthentic representation of myself," says the UX designer who’s been using DatingAI.pro regularly for the past three months. “But I’d also screw up big time when it comes to actually meeting in-person.” He believes you can hardly trick someone into a relationship through using an AI app because texting is such a small part of building that spark – you have to meet them or speak to them at some point.

"The app is more about effective initial exchanges and assisting me to be my best self," adds Artem. "Each user is still making the decisions about what to send and what not to send. For every line I use, there might be 12 that I throw away."

Are dating AI assistants any good?

So how successful is it, actually? Dhruv claims he’s already seen drastic results: "I used to receive one to two matches per week, but after using DatingAI I’ve started receiving six or seven." This success rate is even higher than the hefty promise WingAI makes of increasing your reply rate by 250 percent. Artem feels like his success has stayed the same, but the app has grown his confidence and lowered his stress levels. He sees no shame in using the app, either. "I'm 35, so when I was in high school dating apps didn’t exist and people meeting on the internet was a huge stigma," he continues. "But now the internet and dating on the internet is our default. I don’t see why these apps wouldn’t become a normal part of the process."

But what do daters who’ve never heard of these apps think of all this? London-based Talia, 31, thought it sounded like a slippery slope in the already confusing dating game at first. "If this becomes the norm, surely it means so much extra time wasted going on dates only to find they’re nothing like they came across in their messages?" she says. After more consideration, though, she was thrilled at the prospect of more interesting conversations, but actually testing the apps left her disappointed. "If they gave genuinely good suggestions, I’d be all over it. 

I find myself in Talia’s camp. Frankly, I’m a convert to the actual concept in theory. But the current offerings just don’t hit well enough yet. There’s the price point to consider, too: YourMoveAI is £4.99 per week (scaling cheaper per month and year), WingAI and Rizz are flat fees of £6.99 and a whopping £9.99 per week, respectively.

SEE ALSO: The best dating apps and sites in August 2024

And the biggest point of all still looms large: asking a non-sentient computer program that isn’t capable of love or sex is simply not the same as asking a friend you know and trust, a real-life person who’s actually been on dates before. Plus, could these apps be teaching a whole generation of men with poor flirting or social skills that it’s normal to start off conversations with such inappropriate sexual innuendos? And it is mostly men. Rizz members are currently 66 percent male, while YourMove.AI’s rises to 85 percent and DatingAI.pro a whopping 99 percent.

If the apps can get cheaper and better, they could change the dating game. But more women, and perhaps young people who actually have game, need to be brought in to help programme the software. What do women want? Well, it’s certainly not to be asked about third nipples.

Need more proof that Elon Musk's DOGE team, apparently in charge of making the U.S. government more efficient, has no idea what they're doing?

Here goes: The DOGE.gov website, created to document the findings and changes implemented by DOGE, can apparently be edited by anyone, as the database access has been left wide open.

The first to report this was 404Media, who spoke with two unnamed security researchers confirming the security flaw — if it indeed can be called that, as it appears that there's no "security" of any kind implemented here.

SEE ALSO: Cybersecurity group sues DOGE over data access

As of this writing, two pages on the DOGE.gov site have been defaced, with one saying "THESE "EXPERTS" LEFT THEIR DATABASE OPEN - roro" and the other "This is a joke of a .gov site."

Sigh. Credit: DOGE.gov/unknown hacker

404Media's sources say that the DOGE.gov site was seemingly built on a Cloudflare Pages site that's not hosted on government servers. The examples listed above were created by one of the outlet's sources as to illustrate how insecure the site is.

"Feels like it was completely slapped together. Tons of errors and details leaked in the page source code," they said.

DOGE is a new initiative, started by Trump's administration and led by Musk, tasked to reduce inefficiencies and modernize the federal government. Despite the complex and delicate nature of this task, DOGE has apparently been staffed by college grads with little to no government experience.

Just a few days ago, the newly created Waste.gov site was hidden behind a password after going live with a default WordPress template. Great going so far, dudes.

The National Park Service has removed all mentions of the words "transgender" and "queer" from its web page dedicated to the Stonewall National Monument, a site that marks a landmark moment for LGBTQ rights quite literally led by transgender and gender non-conforming people.

The site's online hub for the New York visitor centre and park honouring the 1969 Stonewall Uprising, a turning point for LGBTQ activism and civil rights in America, now only makes reference to "LGB civil rights," omitting the T and Q representing transgender and queer people in the acronym.

The changes come after anti-LGBTQ executive orders signed by President Donald Trump upon his inauguration, one stating that the federal government now only recognises two sexes, male and female, which medical and legal experts have criticised for biological inaccuracy and ignorance of sexual and gender diversity, and erasure of transgender, nonbinary, and intersex people in America. The Trump administration has been systematically erasing LGBTQ mentions and content from the White House website and federal agency websites.

SEE ALSO: Resources for people worried about Trump's anti-LGBTQ executive orders

Using the Internet Archive's Wayback Machine, Mashable can see the NPS' Stonewall National Monument site has been updated within the last two days.

Here's what it used to look like on Feb. 12:

Credit: Mashable screenshot: National Park Service

And here's what it looks like after Feb. 13:

Credit: Mashable screenshot: National Park Service

The removal of mentions of transgender and queer people from the NPS' Stonewall National Monument website marks a dark, infuriating, unjust day for LGBTQ rights and representation in America. It's a cruel, calculated move that quite literally erases transgender and queer people from a site dedicated to a historical moment of resistance led by trans people of colour and genderfluid people including Marsha P. Johnson and Sylvia Rivera.

"This blatant act of erasure not only distorts the truth of our history, but it also dishonors the immense contributions of transgender individuals — especially transgender women of color — who were at the forefront of the Stonewall Riots and the broader fight for LGBTQ+ rights," said organisers at NYC non-profit Stonewall Inn Gives Back Initiative and the Stonewall Inn (which is physically part of the National Monument) in a statement.

"This decision to erase the word ‘transgender’ is a deliberate attempt to erase our history and marginalize the very people who paved the way for many victories we have achieved as a community. It is a direct attack on transgender people, especially transgender women of color, who continue to face violence, discrimination, and erasure at every turn."

"This decision to erase the word ‘transgender’ is a deliberate attempt to erase our history and marginalize the very people who paved the way for many victories we have achieved as a community." - Stonewall Inn Gives Back Initiative and the Stonewall Inn

GLAAD also published a statement, describing the decision as "yet another example of the Trump administration’s blatant attempts to discriminate against and erase the legacies of transgender and queer Americans."

"The Stonewall Uprising – a monumental moment in the fight for LGBTQ rights – would not have happened without the leadership of transgender and gender non-conforming people," the statement continued. "The tireless work of Marsha P. Johnson, Sylvia Rivera, and countless other trans women of color paved the way and continue to inspire us. You can try to erase our history, but we will never forget those who came before us and we will continue to fight for all those who will come after us.”

New York Governor Kathy Hochul took to Twitter to call the move "cruel and pettty."

Tweet may have been deleted

"Transgender people play a critical role in the fight for LGBTQ+ rights — and New York will never allow their contributions to be erased," she wrote.

It’s a Valentine’s Day Special!

Yellowjackets Melanie Lynskey and Sophie Nélisse reveal the depths of Shauna's rage and how much deeper it gets in Season 3.

The first two episodes ofYellowjackets Season 3 are now streaming on Paramount+ with Showtime, with new episodes weekly.

Last summer, Tinder announced Photo Selector, an AI tool to help pick out the best pictures for your dating app profile. But privacy experts warn that there may be risks associated with the tool.

Photo Selector works by taking a photo of yourself for facial recognition and allowing Tinder to see your photo roll. If you consent to this feature, Tinder gets access to your biometric data (unique physical characteristics in the selfie), and access to your on-device photos. What happens next?

SEE ALSO: Beware of AI tools being advertised on Facebook. They could be malware in disguise. Tinder's use of biometric (and other) data

The good news is, there are some privacy positives about this feature. One, as Tinder explains in its Photo Selector FAQ, is that the feature works on-device. This means that it doesn't take your photos to an external or cloud drive.

As for your biometric data, Tinder states in the FAQ that, "Tinder doesn't collect, store, access or otherwise receive any biometrics generated from your selfie video, profile photo, or photos on your camera roll. Instead, everything takes place entirely on your device, and all biometric data used as part of this feature is deleted from your device once you exit the feature."

Tinder also doesn't collect (store) all the photos on your roll. Rather, it only collects photos that you choose to put on your profile.

Rory Mir, associate director of community organizing at the Electronic Frontier Foundation (EFF), told Mashable that the deletion of biometric data after use does mitigate risks associated with sharing this data. Still, dating app users — and people online in general — should always be cautious when sharing sensitive data like this, Mir said.

"You only have one face," they continued. "You don't want your face print to wind up in a server permanently and [be] used against you in another context." 

Sharing biometrics is currently an opt-in, consent-driven feature at Tinder. The EFF fights for strong privacy protections like this in the private business sector: The ability for consumers to choose to opt-in, that the collection of this data is limited for its intended purpose, and that users can withdraw at any time. (EFF also fights for government use to be fully banned.)

Mir questioned whether Tinder collected metadata or telemetry data, as it wasn't clear in its Privacy Policy. Metadata is "data about the data" — like how many photos are on someone's camera roll — while telemetry data is that of the processes of the app and how it runs. 

A Tinder spokesperson told Mashable that it doesn't collect telemetry data on users using the Photo Selector tool. In terms of metadata, the app collects "limited analytics data." An example the spokesperson gave was that Tinder collects data about how long the tool took to suggest photos, and how many photos were recommended. 

Another general risk Mir identified was the normalization of sharing biometrics. They advised people to avoid sharing biometrics, which would mean avoiding using this feature (and Tinder photo verification using a video selfie). 

What about the Photo Selector AI?

Field chief privacy officer at privacy tech platform Transcend — and former privacy program manager at Tinder — Ron De Jesus said it was notable that Tinder didn't mention whether it trained its Photo Selector AI on user photos in its FAQ.

"There's no mention of how user personal data might be used to enhance or train that AI that's supporting the Photo Selector tool," De Jesus told Mashable. 

In the age of AI, there are concerns that big tech companies like Meta may utilize customer data to train AI models, thus producing generative content based on what humans have put online.

Tinder's spokesperson told Mashable that user photos are not used to train Photo Selector or the algorithms that power it, and that photos are recommended based on Tinder's proprietary algorithms.

This is also good news, especially in the dating app space. Singles share a wealth of their personal data when they use these apps, and sometimes their data is at risk — like when Bumble, Hinge, and other apps had to patch a location vulnerability. Ultimately, dating apps have a great responsibility because of the amount and type of sensitive data they hold, De Jesus said. It seems that, at present, Tinder is mitigating these risks.

UPDATE: Aug. 7, 2024, 2:11 p.m. EDT This story has been corrected from its original version to reflect Ron De Jesus's former title at Tinder.

The Mini is a bite-sized version of The New York Times' revered daily crossword. While the crossword is a lengthier experience that requires both knowledge and patience to complete, The Mini is an entirely different vibe.

With only a handful of clues to answer, the daily puzzle doubles as a speed-running test for many who play it.

So, when a tricky clue disrupts a player's flow, it can be frustrating! If you find yourself stumped playing The Mini — much like with Wordle and Connections — we have you covered.

SEE ALSO: Mahjong, Sudoku, free crossword, and more: Play games on Mashable

Here are the clues and answers to NYT's The Mini for Friday, February 14, 2025:

AcrossWhat each knuckle represents in a famous mnemonic device

• The answer is Month.

Best Picture nominee starring Mikey Madison

• The answer is Anora.

Something set on a microwave

• The answer is Timer.

"Here's to love!," e.g.

• The answer is Toast.

Burnt ___ (barbecue tidbits)

• The answer is Ends.

DownNot glossy, as a photo

• The answer is Matte.

Publication with the headline "History Sighs, Repeats Itself," with "The"

• The answer is Onion.

Wanderer

• The answer is Nomad.

Lock of hair

• The answer is Tress.

Comedian Kevin (with a name that's fitting for Valentine's Day)

• The answer is Hart.

If you're looking for more puzzles, Mashable's got games now! Check out our games hub for Mahjong, Sudoku, free crossword, and more.

Featured Video For You The Wordle Strategy used by the New York Times' Head of Games

Are you also playing NYT Strands? See hints and answers for today's Strands.

Not the day you're after? Here's the solution to yesterday's Mini Crossword.

Alessio Figalli studies optimal transport, a field of math that ranges from the movements of clouds to the workings of chatbots.