TODAY’S PAPER | September 11, 2025 | EPAPER

Well, Technically...

The Express Tribune's science, tech and non-human development stream


Web Desk September 11, 2025 5 min read
Source: Gemini

Another app bites the dust

A terrible day for Android users: Microsoft’s pulling the plug on Outlook Lite on October 6th, ending the app’s short 3 year lifespan.

They’ve been chopping services like this for a while now, with the slated demise of the Lens scanner app in 5 days and the removal of the Editor add-on for Edge and Chrome on October 31st (though that one is going to be built into the browsers themselves).

Microsoft want people to transition to Outlook Mobile instead, as they’re pouring all their investments into that now. I can’t say I care too much about any of this, I have an Android and I don’t think I’ve used Outlook in more than a month. If this affects you, I’m sorry, I’ll play a small violin for you.

AI think, therefore I A.M.

It seems that humanity’s learned absolutely nothing from the Terminator or Matrix movies: Chinese scientists have designed an AI that thinks like a human, mimicking the way neurons fire when humans read sentences and put together their meaning.

And it doesn’t stop there, because apparently, this method of AI reasoning is faster, doesn’t require Nvidia GPUs, takes much less energy, and can be a 100 times faster than current AI models.

Read: Australia to spend $1.1b on 'ghost sharks' - autonomous undersea striker vehicles

So either this revolutionizes the industry and we get fast-tracked to living under the Allied Mastercomputer (A.M, for short), or this fails and we end up paying through our teeth for increased energy bills (no, i’m not biased in my AI coverage, why would you think that?).

Smart home hacking

For all you living in your new-age smart homes, AI prompts can be used to bypass antivirus programs and give connected AI orders with you being none the wiser.

All somebody has to do is slip a prompt in an everyday message that an AI would pick up, and they can make it do things like open a ‘smart’ window, turn on appliances, even reveal your location from your phone, and there isn’t a firewall out there that can stop it.

Not all is lost, however. By making sure you’re up to date on the latest software, monitoring what messages the AI reads, or simply disabling it in places where you don’t absolutely need it, you can prevent this problem from occurring.

Or, you know, just do things manually. That would help too.

Endless? Not quite. Google defines limits on Gemini

Only very recently, we learned what usage limits were actually placed on Gemini at the various tiers. No more guesswork with vague “limited access” disclaimers and the like.

Google has at last updated its Help Center article detailing “Gemini Apps limits & upgrades for Google AI subscribers.” But hold on to your wallets.

That means, up to 100 images per day and 5 prompts per day for free, compared to the 1000 images per day and the 500 prompts per day with ‘Google AI Ultra’. (Screenshot attached above & reviews below).



“Dad, why does the frog have six legs?”

The people over at OpenAI have decided to do away with animators and rely on their AI models to make a full-length animated film: Critterz.

Why deal with those pesky humans with their ‘work-life balance’ and their ‘basic needs’ when you can (almost) remove them from the equation?

Read more: OpenAI to invest $10bn in developing its own AI chips

I say almost because Chad Nelson, the man behind the project, has stated that the animals will be based on sketches that will be provided by people to the AI, and the script and voiceovers will also be done by humans.

So, unfortunately for the OpenAI folks, they will have to deal with people for a little while longer, until they come out with fully functioning androids (so not too long).

Look Mom, I’m on Mars!

 

Whatsapp now lets you generate any custom AI background you could want for your video calls. This is not a new feature, Zoom, and Google Meet, have this feature. But it adds some quality of life to the most used messaging service in the world.

Before you could only use a few basic, boring preset backgrounds, but with this update, you can add a purple dog eating chicken tikka in the background if you want.

Or have your next family call from the surface of the moon.

Also read: Chinese tech giants pursue Nvidia AI chips despite Beijing’s warning

Of course, this is beneficial to people who are worried about their privacy, or the presentation of their environment. You can use simple gradients, solid colors, or even blur your background entirely.

But we used the feature to tell our editor that we’re at the beach today. Whatever you think about it, at least it’s free (looking at you, Snapchat).


Is Google Photos + Veo a good match?

Google’s ‘best’ video-generation model, Veo 3 is on Google Photos now. But the question is: does this even matter to the average user?

The new model, available on the mobile app’s Create tab, will allow users in the US to turn their still images into short video clips. Google Photos already offers video generation through its recently added “Photo to video” feature, but Google claims Veo 3 enhances that functionality with higher-quality videos.

But how significant is the difference? And what about all the Google Photos users outside of the US?

“Hey Lens: Imagine… Dragons”


So Snapchat’s unveiled their newest cash grab feature: the Imagine Lens. With this, you can create AI images with text prompts, letting you dream up whatever wacky, zany scenario you want to put yourself in. Make you a cowboy? Done. Riding a dragon? You got it.

But as with any good or even halfway mediocre thing these days, it’s paywalled behind either a Snapchat + platinum or Lens+ subscription, meaning you have to cough up $9USD a month if you want to spice up your streaks.

Only time can tell if people will use this feature regularly, or if they’ll remember it exists while swiping through the options.

COMMENTS

Replying to X

Comments are moderated and generally will be posted if they are on-topic and not abusive.

For more information, please see our Comments FAQ