Bing's AI Chatbot Innovates as Samsung Sticks with Google

Today:

Bing’s AI chatbot can make charts now

Bing's chatty AI bot has learned a new trick: making charts! According to the release notes, it can whip up some fancy visualizations. But hold your horses before you start praising its math skills. You might want to give those numbers a good ol' double-check just to be safe. Better safe than sorry, folks!

READ THE ARTICLE ON THE VERGE.

Samsung’s sticking with Google as its default mobile search engine

Well folks, despite some flirty side glances at Bing, Samsung has decided to stick to its tried-and-true sweetheart, Google, for its default mobile search engine.

According to the grapevine, aka The Wall Street Journal, Samsung had been toying with the idea of giving Bing a whirl on its own web browser. That got some eyebrows raised over at Google, who weren't thrilled about potentially losing the cool $3 billion a year they pocket from this deal.

It seems like Samsung figured, "Eh, most of our customers barely touch our in-house browser, how big of a kerfuffle could this really cause?" But then it seems they took a beat, maybe sipped on some common sense juice, and realized this switcheroo could ruffle some feathers with Google and cause some market mumbo jumbo.

But don't take this as the end of the story. Samsung isn't exactly slamming the door on Bing's face, just maybe telling it "It's not you, it's me." There's always a chance they'll give Bing a shot down the line. But for now, Google's still their main squeeze.

READ THE ARTICLE ON THE VERGE.

Labour should pledge £11bn to build ‘BritGPT’ AI, thinktank says

So, here's the skinny. This group of big thinkers called "Labour for the Long Term" is banging the drum for Keir Starmer and the Labour party to cough up a whopping £11bn to create a British-branded AI system, creatively named "BritGPT", and a cloud to boot. They're pushing this hard, afraid that if they don't, Britain will end up as the awkward third wheel to American tech giants.

Their big worry? Britain becoming a one-hit wonder if we're overly reliant on the likes of Google and Amazon. They reckon the government's current offer of £1bn is a bit like bringing a pea-shooter to a bazooka fight, given the stakes. Their idea of a solution? Build a homegrown AI powerhouse to focus on fixing market failures, not just trying to out-muscle Silicon Valley.

Their argument is that private companies, always with an eye on the bottom line, won't cough up enough dough for 'AI for good' or to keep AI safe. They propose the UK government fill in that gap, with a special focus on things that benefit everyone like medical and clean energy research. The suggestion is that they could even dip into the £28bn climate investment pledge to find the cash.

Haydn Belfield, an egghead over at Cambridge University, stresses that AI is like the new cool kid in the technology schoolyard and the UK better make friends fast or risk getting left behind. He's pretty vocal that the government's current investment into AI training and building a super-duper computer are like trying to fill a bathtub with a teacup - not nearly enough.

According to Belfield, the kind of money that heavyweights like Amazon or Google sink into their services is closer to £10bn, and we can expect the cost of future AI systems like GPT-5 or GPT-6 to rocket into the hundreds of millions, not even counting the cost of the smart folks you'd need to hire. He's advocating for a £10bn "Great British cloud", which is right in line with Labour's other patriotic proposals.

And it's not just these guys sounding the alarm. Geoffrey Hinton, the brain behind "deep learning", warned that AI development could turn into a big, scary monster if driven only by dollar signs. He pointed out that Google, despite being at the front of the pack in AI tech, has kept some of their goodies under wraps, not wanting to spoil their good name. But the issue, as he sees it, is that in a capitalist world, if a rival spills the beans, you have no choice but to follow suit.

READ THE ARTICLE ON THE GUARDIAN.

How the digital ad industry can guide the ways AI transforms businesses

Alright folks, listen up! You know how OpenAI launched that smarty-pants AI, ChatGPT, back in February? Well, it's pretty much rocked our socks off and made a bunch of us rethink our life choices, right down to our Bing search habits. Now everyone and their grandmas are chattering about big words like 'large language models' and 'natural language processing.' And it's not just the tech whiz kids who are paying attention - even bigwigs and rule-makers are raising eyebrows at all the hoopla around AI and privacy.

Oh, and you know that chap, Elon Musk? His posse at the Future of Life Institute got a cool 1,000 signatures to hit the pause button on training AIs fancier than GPT-4. Makes you wonder: if tech boffins take a break from the AI race, will they spend their summer vacations figuring out how to make AI play nice with privacy and ethics?

And while everyone from Google and IBM to Amazon and Baidu are jumping on the AI bandwagon, there's still a ton we don't know about what's around the corner. That's why we gotta buckle down on privacy, ASAP - because nobody likes nasty surprises.

As for the ad industry? Well, they're rubbing their hands together, eager as beavers to put AI to work. But they've got to watch their step - things like personal info, discrimination, data sharing, and worldwide regulations are all in play. It's like walking a tightrope while juggling flaming swords.

And let's talk about the looming showdown between search and AI. Imagine how easy life could be with all the info you need gathered by AI, instead of doing the digital equivalent of chasing your tail with search queries. If the kiddos embrace AI as the new normal, search engines could be on their way to join the dinosaurs, and the ad industry will be in a bit of a pickle.

But let's not forget the publishers, folks. They still rely heavily on search for traffic, and with ads sneaking their way into AI chat, they're asking the million-dollar question: how will they get their piece of the pie? They'll need to be savvy to navigate the minefield that's coming their way.

Which brings us to the big issue: privacy. The big guns need to be extra careful about how they handle data for AI. That includes getting clear consent from users before collecting their data and making sure users can opt out if they want to. Convenience is great and all, but not at the cost of privacy. That's a bill nobody wants to foot.

And let's not forget, the spotlight is on the big guys now, but with all the brainpower being dispersed from layoffs, don't be surprised if the next big AI breakthrough comes from the little guys. So, publishers and industry leaders, this is your wake-up call. It's time to gear up and get ready for a future that's privacy-safe, transparent, and profitable.

Fred Marthoz, the VP of global partnerships and revenue at Lotame, couldn't have said it better. But remember folks, we're all in this together, and the future is what we make of it.

READ THE ARTICLE ON VENTUREBEAT.

The thing missing from generative AI is the ‘why’

Alright, so here's the skinny. Generative AI, the type of tech that companies like Google and Meta are jazzed about, can cook up ads and test them, making small businesses sing. Imagine creating brand-friendly, social-media-ready ads without breaking a sweat.

But the big brands? They may be in a pickle. Why? Because AI doesn't really get the 'why'. It's a wiz at creating and testing ads but it doesn't understand why one ad does better than the other.

Say you've got two ads for a fast food joint. One's a hit, the other's not. The AI sees the hit ad's got a burnt orange color and thinks, "Bingo! Burnt orange is the secret sauce!" But a human peeps that the burnt orange is just coffee with cream. The cream's the magic, not the color. See the mess?

And if you think that's complicated, brands also need to decide whether to show folks like you and me or go all Hollywood with their models. This isn't just about what sells, but about reflecting the real world in ads. Can AI understand and respect this delicate balance? That's a tough one.

To put it in a nutshell, AI is good for taking the grunt work out of creating and testing ads. It can't replace the good ol' human brain when it comes to understanding 'why' an ad works and making brand-sensitive decisions. That's where us humans still shine. But if we can teach AI these nuances, it can be a powerful sidekick to brands, creatives, and data nerds. So, it's not a lost cause, but there's work to be done.

READ THE ARTICLE ON VENTUREBEAT.

Wisconsin Police Department Warns of New Artificial Intelligence Phone Scam

Alright, buckle up, folks in cheese land. Some cheeky swindlers are pulling a fast one, using your grandma's voice to scam you out of your hard-earned dough.

The boys in blue from the Beloit Police Department have noticed this latest trick. They're not sure if it's fancy AI stuff making this possible, but they want you wise to it. Basically, they make a clone of your dear old aunt's voice, ask you for some cash because they're "in trouble," and you're none the wiser.

This ain't small potatoes, either. Even the bigwigs at the Senate Special Committee on Aging are rattled, hitting up the Federal Trade Commission to see how they're keeping grannies safe from these shenanigans.

Now, you're probably thinking, "this sounds tougher to pull off than a one-legged man at a butt-kicking contest," but it ain't. All these crooks need is a recording of your cousin's voice and some off-the-shelf software.

The play goes like this: you get a call from "cousin Jimmy" who's been nabbed by the cops, or gotten his fool self in a car wreck, and needs a quick cash injection.

So how do you keep from getting taken for a ride? Easy peasy. Ring up "Jimmy" on a number you know is his, not the one he called from. If he doesn't answer, ask around the family, see if his story checks out.

And remember, if "Jimmy" suddenly has a burning need for wired money, Bitcoin, or a bunch of gift cards, that's as fishy as a three-dollar bill. Probably a scam.

To keep this from happening, go through your social media and make sure you're not giving out your life story to all and sundry. These scammers need fodder to seem legit, so don't make it easy for 'em. Stay sharp, and keep your wallet safe!

READ THE ARTICLE ON NBC.