Internet's Best Secrets

The origin of the Nano Banana name

2026-01-21T12:49:00.000+00:00

The popular Nano Banana AI image generation model has a curious origin story regarding its uncommon name.

Google DeepMind has explained the unexpected origin of one of its most viral AI names: Nano Banana. Officially known as Gemini 2.5 Flash Image, the model quickly became famous not just for its image-editing abilities, but also for a codename that stood out in a world usually dominated by dry, technical labels.

The name wasn’t planned by marketing. It was created at 2:30 a.m. as a last-minute placeholder for LMArena, a platform where AI models are tested anonymously. Product manager Naina Raisinghani combined two personal nicknames - “Naina Banana” and “Nano” - to create something memorable. The team knew the name was silly, but kept it because the model was a fast, lightweight “Flash” variant, making “Nano” oddly fitting.

Once Nano Banana appeared publicly, the AI community took notice. Alongside the unusual name, users were impressed by the model’s ability to maintain character consistency, restore old photos, and handle culturally specific image prompts. As interest spread on social media, the codename became inseparable from the model’s identity.

Instead of reverting to a formal name at launch, Google embraced the joke. Banana-themed UI tweaks, emojis, and even branded swag followed. With the release of Gemini 3 Pro Image, the branding has now gone official with Nano Banana Pro, proving that a spontaneous late-night decision can turn into a lasting and surprisingly effective AI brand.

Google Forms gain response limits by date or number

2026-01-14T13:36:00.009+00:00

Google Forms can now stop collecting data at a specific date or whem it reaches a preset number of submissions.

Google has finally added one of the most requested features to Google Forms: automatic response limits. Instead of manually turning forms on and off, creators can now set clear rules that decide exactly when a form should stop accepting submissions.

There are two new automation options. The first lets form owners choose a specific date and time for responses to close, which is ideal for registrations, applications, or surveys with strict deadlines. Once that time is reached, the form automatically closes and can show a custom message to anyone who tries to respond late.

The second option allows creators to cap the total number of responses. As soon as the limit is reached, the form shuts down on its own. This helps avoid issues like overbooking events or collecting more responses than intended, without the need for constant monitoring.

The feature is rolling out to all users, including Google Workspace customers, Workspace Individual subscribers, and even personal Google accounts. It’s turned off by default and can be enabled from the Responses tab after publishing a form. The rollout began on January 12, 2026, and may take up to 15 days to reach everyone.

Tesla add telemetry data to video recordings

2025-12-10T11:57:00.004+00:00

Tesla is finally giving vehicle owners access to telemetry data that was previously locked away inside the company’s proprietary logs.

With the 2025 Holiday Update, alongside showing the pedals state in the recorded videos, Tesla has added basic telemetry to the in-car Dashcam interface and released a new Dashcam Tools repository on GitHub, allowing owners to decode and analyze the data embedded in their dashcam footage. It’s effectively the first time Tesla drivers have been handed the keys to their own "black box".

The highlight is a browser-based viewer where users can drag and drop a dashcam file to see synchronized, frame-level telemetry. The video now contains embedded SEI (Supplemental Enhancement Information) metadata showing real-time accelerator position, brake usage, steering angle, Autopilot/FSD state, turn signals, GPS coordinates, acceleration vectors, and more. For accidents, insurance claims, and legal disputes, this turns subjective interpretation into verifiable numbers extracted directly from the video.

This new capability is enabled by Tesla’s updated video encoding introduced in firmware 2025.44.25, which merges telemetry with visual frames for perfect synchronization. However, the feature requires vehicles equipped with HW3 or HW4, as only those systems can handle the additional processing overhead. Tesla also notes that Sentry Mode recordings may lack telemetry, since parked vehicles power down many sensors to conserve energy.

Despite the viewer running in a web browser, Tesla says all processing happens locally, meaning no footage is uploaded to Tesla or any external server — an important safeguard for privacy-sensitive material. For deeper inspections, Tesla’s GitHub repository also includes a metadata extraction tool. Together, these changes give owners unprecedented insight into their car’s behavior and a far more transparent set of tools for understanding incidents on the road.

Tecno unveils Freeform Continuum Telephoto and Dual-Mirror Reflect Telephoto lens for smartphones

2025-12-04T13:23:00.003+00:00

Tecno has unveiled its new Freeform Continuum Telephoto and Dual-Mirror Reflect Telephoto lens assemblies for smartphones.

Tecno used its 5th annual Future Lens event to showcase two major mobile camera innovations: Freeform Continuum Telephoto and Dual-Mirror Reflect Telephoto. The company says both technologies aim to push smartphone zoom capabilities far beyond today’s fixed-focal-length telephoto systems.

The Freeform Continuum Telephoto module is designed as a single periscope camera that delivers smooth, continuous optical zoom from 1x to 9x. Because it serves as both the main camera and the telephoto lens, it eliminates the typical jump and color shift that occur when phones switch between multiple rear cameras. Tecno claims this approach also delivers higher image quality at mid-range zoom levels than traditional digital zoom.

Tecno also unveiled the Dual-Mirror Reflect Telephoto, a compact module that is 50% smaller than standard lens-based telephoto designs. Instead of stacking lenses, it relies on two mirrors — similar to classic mirror lenses or miniature catadioptric telescopes. This system produces a distinctive donut-shaped “iris blur” bokeh effect, which some users may see as a creative advantage. The trade-off is a one-stop loss of light, though Tecno says the technology is production-ready and could appear in phones as early as next year.

The more advanced Freeform Continuum system is still under development and at least a year away from commercial release. Tecno has already secured Samsung and Largan as manufacturing partners, signaling that both technologies could soon make their way into future high-end smartphones.

Runway Gen-4.5 overtakes Google Veo 3

2025-12-02T13:15:00.010+00:00

Runway is back to the top with its latest Gen-4.5 AI Video model, surpassing Google's Veo 3.

Runway has officially shaken up the video AI landscape with Gen 4.5, a model developed by a team of around 100 researchers that has unexpectedly overtaken both Google and OpenAI on the industry’s leading blind video benchmark. The system uses a new Autoregressive-to-Diffusion (A2D) architecture that blends the logical sequencing strengths of autoregressive models with the visual fidelity of diffusion, giving Runway a major leap in realism, object permanence, and motion physics.

This hybrid approach allows the model to understand scenes through a vision-language backbone while rendering frames in parallel, avoiding the typical flickering and distortion seen in competing models. As a result, Gen 4.5 maintains consistent lighting, fluid dynamics, and character movement, addressing long-standing weaknesses in generative video. Benchmarks back this up: on the Video Arena leaderboard, Gen 4.5 leads with an Elo score of 1247, ahead of Google’s Veo 3, Kuaishou’s Kling 2.5, and even OpenAI’s Sora 2, which has slipped to seventh place.

Introducing our new frontier video model, Runway Gen-4.5. Previously known as Whisper Thunder (aka) David.

Gen-4.5 is state-of-the-art and sets a new standard for video generation motion quality, prompt adherence and visual fidelity.

Learn more below. pic.twitter.com/2J945S4bQt
— Runway (@runwayml) December 1, 2025

Gen-4.5 achieves an unprecedented level of physical accuracy and visual precision. Objects move with realistic weight and momentum. Surfaces behave the way they would in the real world. Laws of physics can be both observed or ignored, depending on your vision.

Despite the big… pic.twitter.com/AsbLQ0ZSDH
— Runway (@runwayml) December 1, 2025

The model isn’t perfect — notably, it lacks built-in audio generation, leaving creators to rely on external tools for dialogue, sound effects, or ambience. That gives Google’s Veo an advantage in convenience, even if Runway currently dominates on visual quality and prompt accuracy. Still, the platform includes useful style presets such as Photorealistic, Cinematic, and Slice of Life, and it demonstrates surprising improvements in physics and object permanence that set a new standard for the field.

Looking back at how things were just two years ago, when Runway Gen-1 was launched, it makes us wonder how things will be two years from now.

Two years ago, we introduced Gen-1, the first publicly available video generation model. It enabled an entirely new form of creative expression and created a new product category.

Over the past two years, we've led the industry in making video models more powerful and… pic.twitter.com/eLJfFtvQUg
— Runway (@runwayml) December 1, 2025

Runway Gen 4.5 pricing reflects the high cost of running on NVIDIA H100 and Blackwell GPUs. Plans start with a free tier offering 125 credits (enough for short tests), while paid tiers scale up to the $76/month Unlimited plan, which is positioned as the best option for heavy creators thanks to unlimited, throttled rendering. With its strong technical leap and clear win over Google Veo 3 in blind evaluations, Gen 4.5 solidifies Runway as a serious challenger to the biggest players in AI video.

Netflix removes casting from app to TV

2025-12-01T13:41:00.002+00:00

Netflix "enshitification" continues, now removing the ability to cast from a smartphone to TV.

Netflix has quietly removed the ability to cast content from its mobile app to most TVs and streaming devices, catching many users by surprise. Over the past few weeks, people noticed the Cast button disappearing from the app, and Netflix has now confirmed the change in its updated support pages. The shift affects popular devices like Chromecast with Google TV and Google TV Streamer, regardless of which subscription plan you have.

According to Netflix, casting is no longer supported on TVs or streaming devices that come with their own remote. Instead, users must navigate Netflix directly on the device. The company claims the decision is meant to “improve the customer experience,” though many users disagree, especially those who preferred the convenience of using their phone to browse and control playback. This update also wipes out the old exception that allowed casting on newer Chromecast with Google TV models if you were on the ad-supported Standard plan. Netflix has now made it clear that casting from a mobile device won’t work on most modern devices, even on the fully paid, ad-free tiers.

The only remaining option is older Chromecast models without remotes and TVs that still support native Google Cast. Even then, casting is restricted: ad-supported plans cannot cast at all, while ad-free plans continue to work on these legacy devices. Reports from users echo the same situation - traditional Chromecast still shows up in the app, but anything running Google TV no longer does.

Google Calendar updates flairs for months and new types of events

2024-09-29T15:21:00.003+01:00

Google Calendar is preparing new flairs for event types, as well as updated month imagery.

Google Calendar have long known there are specific keywords that can automatically add images to their events. And soon, we'll get some new ones.

References found inside the latest Google Calendar app reveal there are some new flairs incoming, for things like train rides, as well as updated imagery for anniversaries and months.

The complete list of keywords is quite extense and can be found GitHub for those who are curious enough, or are simply looking for a way to make their Google Calendar events become visually more distintive than a simple grid of blocks.

How to get old-style compact view on the new Blogger interface

2020-10-09T11:37:00.000+01:00

Google has decided to finally revamp its nearly forgotten Blogger interface, but sadly the result is far from perfect. Thankfully, there's a way to make it look a lot more like the previous Blogger, with the help of some tweaked CSS.

Though there are a lot of bugs, missing features, and assorted annoyances, related to the new blogger interface (truth be told, the blogger team is doing its best to fix most of the issues), the first and most annoying thing blogger users notice is the new, dreaded, post list that went from being a list to a "card view" that wastes lots of space.

Even on full-screen, you'll likely get under 10 posts visible on the list, which isn't nearly enough for higher volume blogs and, worse of all, it makes a lot harder to have a sense of the posts dates, published / scheduled status, etc. But with the help of some tweaked CSS, you can get it to look like this:

For that we need to use a browser extension than can apply new CSS rules to existing pages, like Stylebot for Chrome. And after installing it you just need to add the following rule set, slightly modified from this one available on GitHub, for the blogger.com site.

/* New Blogger Version 0.951 aadm */
 
/* Post Editor Label Box */
.KHxj8b {
    color: blue;
    font-size:14px;
    line-height: 20px;
    /* height: 132px !important; */
}

/* Draft Label Colour and Size */
.haXJ6e .htI7vb {
    color: #7a7;
    font-size: 12px;
}

/* Published Label Colour and Size */
.gRzh8c.NY7gZe {
    display: none;
}

/*.NY7gZe {
    color: #333;
    font-size: 12px;
}*/

/* font title bolder and smaller */
.haXJ6e .UHwcef {
    font-size: 14px;
    font-weight: 450;
}

/* Label stuff. Not 100% sure*/
.haXJ6e .qtvRHc {
    padding: 0px 0px;
    font-size: 11px;
    text-align: center;
}

.haXJ6e .NY7gZe {
    font-size: 12px;
}

/* Set Scheduled Color */
.UNTnVe {
    color: rgb(240, 137, 0);
    font-size: 12px;
}

.uZsIBe {
    display: inline-flex;
    flex-wrap: wrap;
    margin-left: 0px;
    margin-right: 0px;
}

/* Label borders and alignment */
.qtvRHc {
     border: 0px solid #b0bec5; 
     border-radius: 0px; 
    text-align: left; 
    color:#2196f3;
}


/* Outer Div for Label List on dashboard */
.Q0PqBe {
    flex: auto;
    overflow: hidden;
    margin-left: 10px;
    line-height: 1;
    margin-top: -12px !important;
    display: block;
    position: relative;
    left: -150px;
}


/* Author Name */
.GLDMfb.UHwcef {
    font-size: 11px !important;
    color: rgb(131, 131, 141);
    font-weight: 200;
}

/* Remove Comment Count */
.jLK47 {
    display: none;
}

/* Remove View Counter Icon */
.rtpUCe.JB36v {
    font-size: 18px;
    display: none;
}

/* View Counter Size and Color */
span.dYZApe.NY7gZe {
    font-size: 11px;
    color: cadetblue;
}

/* Remove Profile Icon */
.qlgjDe {
    display: none;
}

/* Reduces the gap between post and inner padding*/
.oqIa7e {
    margin: 0 0 -6px 0;
    padding: 0px;
}

/* Change the backgroun color to match the row hover color */
.oqIa7e:focus-within .FeZNBc, .oqIa7e:hover .FeZNBc {
    background-image: none;
    background-color: aliceblue;
}

/* Editing Tools better padding and align flush to the right */
.FeZNBc {
    padding-top: 6px;
    right: 0px;
    top: 8px;
}

/* Change from box lines around a row to just top */
.vOSR6b {
    background-color: #fff;
    border-top: 1px solid #e9f0f3;
    border-radius: 0px;
    box-shadow: none;
    border-left: 0px;
    border-right: 0px;
}

/* Highlight the hovered row */
.vOSR6b:hover {
    background-color: aliceblue;
    border-top: 1px solid #e9f0f3;
    border-left: 0px;
    border-right: 0px;
    border-radius: 0px;
}

/* Remove .separator between scheduled text and date */
span.wPZNb {
    display: none;
}

/* Reduce size of hover over tools and color to black */
.DPvwYc {
    font-size: 16px;
    color: black;
}

/* Edit Form Label Row Padding and font size */
.OGidW {
    padding: 8px 4px !important;
    font-size: 16px;
}


/* Increase the Width of the Sidebar in Edit Mode */
.vW7mGd {
    width: 320px;
}


/* Increase the padding of the Editor Content */
.fmzcZd {
    padding-right: 340px;
}


/* Move scheduled over - hard coded */
.ngYWic {
    /* align-self: center; */
    flex: 0 0 auto;
    position: relative;
    right: -85%;
    width: 140px;
}


/* Remove rounded on thumbnails */
.sRb2Nc.DMO9ie {
    border-radius: 0px;
}


/* Reduce the main dashboard container padding and margins */
.n8rJwb {
    padding-left: 0px;
    padding-right: 0px;
    padding-top: 0px;
    margin-left: 0px; 
    margin-right: 0px; 
    max-width: 100%;
}


/* Row Select Tick Box */
.PPvste {
    left: -15px;
}


/* Remove Thumbnail */
.jn15V {
    display: none;
}


/* Reduce the row left pad */
.VSOPnd {
    padding-left: 6px;
}


/* Moves the post title closer to the horizontal row separator */
.Bv5FY {
    margin-top: -6px;
}


/* Select All Left Aligned */
.HJyrF {
    display: flex;
    height: 48px;
    justify-content: flex-start;
}
/* Unopened Labels Text */
.dLSeoc {
    color: cornflowerblue;
    font-style: normal; }

Sure, it doesn't fix evereything, lie all the new editor bugs and quirks; but at least it makes the Blogger blogs main screen manageable once again. I still hope the new Blogger will eventually offer an official compact view, but until then, desktop users hoping to mimick the old blogger look, can resort to this technique to get it right away.

Covid-19 makes Face ID a nuisance - will that make Apple bring back Touch ID for iPhone 12?

2020-07-28T13:47:00.000+01:00

Covid-19 has transformed the habits of our society and, with the widespread (and in some cases mandatory) use of face masks, it also reveals that companies that went all-in on facial recognition systems as a way to unlock smartphones, as is the case with Apple with its Face ID, made the wrong move.

Contrary to what was expected at the time - that Apple would adopt a fingerprint sensor on the screen - Apple chose to adopt only a facial recognition system (Face ID) from the iPhone X onward. Even though it was a technological leap at the time, it does not solve the issue of not working with a face mask on. The convenience of taking an iPhone out and, almost instantly being unlocked and ready to use, disappears, and instead we go back to the days where we need to manually enter a PIN or password.

The use of face masks has launched chaos among facial recognition systems, but if we focus only on the regular smartphone use on a daily basis, it reveals how reckless the bet on facial recognition as the single authentication method was/is. To make matters worst, this happens at a time when the smartphone is used more and more, as a payment method for instance, making the added inconvenience of unlocking it even more burdensome. The ritual of taking an iPhone out and having it ready to make a payment with just a glance of your face (via Face ID) is gone; and you're now back into typing a PIN code, or pulling down the mask so it can recognize you.

Nothing has been heard so far about the possibility of Apple adding a Touch ID on the screen to the iPhone 12, but this seems to be essential for this new post-Covid-19 era. If this year's iPhone 12 comes only with Face ID yet again, and taking into account that the mandatory use of face masks in public spaces should remain in place for a long while, the iPhone 12 unlocking process will continue to be anything but practical. Hopefully, in September, Apple will surprise us by announcing that the iPhone 12 has both Face ID as well as Touch ID integrated on-screen; otherwise... will be paying top dollar for a high-end device that makes us go back nearly a decade to the days of the pre-Touch ID iPhones.

Botnet is a social network where everyone is famous

2020-03-02T13:00:00.000+00:00

Anyone wanting to get a feel for what it's like to be a star online can now do so on the Botnet network, that catch is that the thousands of people liking and commenting your posts are not real.

Botnets are mostly associated with negative purposes, attacks by hackers, but in this case it is a very peaceful and positive botnet. This social network Botnet allows all users to have hundreds of thousands of followers right away, eager to leave likes and positive comments to everything that you post there.

The service is free and allows anyone to join a social network that, even though it is composed entirely of digital interactions with bots, produces real positive effects. As its creator says, when you're dealing with millions of followers, you don't get to know who is real and who is a bot anyway.

You can even unlock some extra features, like double the number os followers, as well as enable (fake) troll comments - an option that seems to be quite popular. Fortunately, they come in a much smaller number than you'd likely face in real life, and which can be turned off at any time (also unlike what would happen in real life).

It is still a fun experience, and it can be the perfect "fix" for those tired of social networks with real people. :)