Hashtag Trending Nov.17-YouTube previews AI tool to clone famous singers; Samsung admits to year-long breach; AI system wrongly rejects patients’ health claims

Share post:

YouTube cracks down on AI-generated deepfakes, then previews an AI tool that clones famous singers, Samsung admits to year-long data breach, X is accused of deceptive ad practices, and AI unlawfully rejects healthcare claims of patients.

Hashtag Trending on Amazon Alexa Google Podcasts badge - 200 px wide


These and more top tech stories on Hashtag Trending

I’m your host, James Roy.

Just days after YouTube unveiled policies to combat AI-generated deepfakes on its platform, it’s now previewing an AI tool that clones famous singers. Right, confusing.

The AI tool is called Dream Track and it will allow creators to make short, AI generated songs, using the voice of artists. It can generate lyrics, a backing track, and an AI-generated voice in the style of the artist.

The tool is powered by a music generation model from Google DeepMind, called Lyria. Google says tracks created using Lyria will carry a SynthID watermark that’s inaudible to the naked ear and can be preserved when a track is modified. So even if someone adds more noise to a track, compresses it into an MP3 file, or speeds it up, it should theoretically still be possible to tell that it contains AI-generated audio from Lyria.

So far, only nine musicians have consented to the project. One artist said they’re still wary, and another said they’re hopeful that AI will accelerate creativity rather than replace it.

One YouTube executive working on generative AI in music, however, resigned, saying he cannot support creating AI tools using an artist’s discography that would ultimately replace them.

YouTube was in talks with all the major labels about the tool, but talks have stalled with Sony Music and Universal.

Source: The Verge, Gizmodo

Apple is barely recovering from all the heat (literally) that it got from the release of the latest iPhone 15. But it’s already sorting out ways to solve overheating issues in its next flagship.

Rumours are swirling that the company is considering switching to a graphene heat sink and a metal battery case to promote better heat transfer.

Graphene can be optimal for a heat sink because it has ten times more thermal conductivity than copper.

Apple blamed a software issue, which it now has fixed, for the major overheating issues in the iPhone 15 series. However, the company said that more effective cooling can also help.

It investigated many ways to cool the iPhone down but hasn’t changed much beyond using different heat sinks, some being cost-prohibitive or some requiring radical design changes.

Graphene may be the perfect solution to keep the iPhone 16 from overheating if the rumour turns out to be true. 

At least then, Apple’s new phone would actually be different from its predecessors…?

Source: Apple Insider 

Samsung has admitted that hackers accessed the personal data of customers who made purchases at Samsung U.K.’s store between July 2019 and June 2020.  

The company said it did not discover the compromise until a few days ago.

Samsung told affected customers that hackers may have accessed their names, phone numbers, postal addresses, and email addresses. But no financial data, such as bank or credit card details or customer passwords, were impacted.

The breach was reported to U.K.’s Information Commissioner’s Office, which said that it’ll be making enquiries.

This incident is the third data breach that Samsung has disclosed in the past two years.

Source: Tech Crunch

Check My Ads, an online advertising watchdog has filed a complaint with the FTC against X for ad malpractices. 

The complaint alleges that the recent changes to the platform obscures whether a post that a user is viewing is an advertisement. 

Previously, advertisements were clearly labeled with a “Promoted” affixed to the bottom of a paid ad.  However, Mashable reported in July that many users stopped seeing the Promoted label on paid posts.

A new “ad” label, placed on the upper-right corners of posts, replaced it — and it was much less noticeable.

But, some ads had no label at all. Mashable reported that the only way users could determine a post was an ad was if they accessed a drop-down menu that provided them with an option to report the ad.

Even worse were clickbaity ads that started rolling out on X. Not only did these ads not disclose their nature but also hid the advertiser behind the ad.

X previously paid a $150 million fine, for deceptively using users’ information for advertising purposes, months before Elon Musk’s takeover. Meanwhile, advertisers have been abandoning the platform, wiping out 60 percent of its advertising revenue earlier this year.

Source: Mashable 

In yesterday’s episode, we unpacked the promises of Forward Health’s Care Pods, a standalone medical station using AI to conduct tests, spit out a diagnosis, prescriptions and so much more at less than $99 a month.

U.S. health insurance giant, UnitedHealthcare has also been using AI but it does not sound like it’s going too well.

The company has been hit by a lawsuit because it allegedly uses a faulty AI that repeatedly and wrongfully refused to pay the healthcare claims of senior patients.

The complaint also says that the AI system has a 90 percent error rate that improperly overrode the recommendations of physicians regarding medically necessary post-acute care for the elderly, i.e. treatment following discharge from a hospital.

Even more concerning, UnitedHealthcare allegedly instructed employees not to deviate from its AI model’s predictions or those who do would be disciplined or terminated.

Ryan Clarkson, managing partner of Clarkson Law who brought the lawsuit said, 

“UnitedHealthcare is responsible for the health care needs of some of our most vulnerable populations. But by invoking this technology, they are effectively using AI to throw the elderly – our parents or grandparents – out onto the street.”

Source: The Register 

And that’s the top tech news for today.

Hashtag Trending goes to air 5 days a week, and there’s a special weekend interview show on Saturdays we call “the Weekend Edition.”

You can get us anywhere you get audio podcasts and there is a copy of the show notes at itworldcanada.com/podcasts 

For those interested in cybersecurity you can also check out our hit cybersecurity podcast featuring Howard Solomon. It’s called CybersecurityToday.  And it’s rated as one of North America’s top 10 tech podcasts. 

I’m your host, James Roy – have a Fantastic Friday!

The post Hashtag Trending Nov.17-YouTube previews AI tool to clone famous singers; Samsung admits to year-long breach; AI system wrongly rejects patients’ health claims first appeared on IT World Canada.


Related articles

Will the “AI bubble” burst? Hashtag Trending for Wednesday, July 10, 2024

Europe may be reigning in big tech, but Canada and the US are struggling, despite public concern.  Analysts...

Who is the real competition for OpenAI? Hashtag Trending for Tuesday, July 9, 2024

Claude 3.5 is giving OpenAI a real run for its money.  It’s a war of the bots –...

Cyber Security Today, July 8, 2024 – New ransomware group discovered, and summer podcast break starts

A new ransomware group is discovered. Welcome to Cyber Security Today. It's Monday July 8th, 2024. I'm Howard Solomon,...

Realizing the costs and impacts of AI and social media: Hashtag Trending in the Summer for Monday, July 8, 2024

Well, I’m back – what did I miss? It seems like we’re discovering the real costs of AI –...

Become a member

New, Relevant Tech Stories. Our article selection is done by industry professionals. Our writers summarize them to give you the key takeaways