Connect with us

Technology

Bias, deaths, autonomous cars: Expert says AI ‘incidents’ will double as Silicon Valley launches tech race

Published

on

Bias, deaths, autonomous cars: Expert says AI ‘incidents’ will double as Silicon Valley launches tech race

As Silicon Valley races to construct highly effective and standard synthetic intelligence programs, troubling “incidents” starting from convincing AI deepfakes, banking fraud, bias and even deaths will enhance this 12 months, a tech professional says.

Following the discharge of ChatGPT final November, tech firms have been speeding to develop highly effective AI programs to maintain the tempo with rivals.

The AI Incident Database, which is run by nonprofit Accountable AI Collaborative, tracks varied incidents brought on by AI and is projected to report double the variety of incidents this 12 months in comparison with final. The database defines incidents by examples comparable to an autonomous automobile killing a pedestrian, a “buying and selling algorithm” inflicting a “market ‘flash crash’ the place billions of {dollars} switch between events,” or a “facial recognition system” inflicting “an harmless individual to be arrested.”

“There isn’t a such factor as a minor incident with AI,” stated Sean McGregor, the founding father of the AI Incident Database who beforehand labored as a machine studying architect.  

“If I produce a system that makes 8 billion folks really feel barely extra depressed, then that system may have pushed some variety of folks to suicide,” McGregor advised Fox Information Digital. “The unlucky matter is that with fantastical scale even small impacts add up.”

Advertisement

REGULATORS SHOULD KEEP THEIR HANDS OFF AI AND FORGET MUSK-BACKED PAUSE: ECONOMIST

Following the discharge of ChatGPT final November, tech firms have been speeding to develop highly effective AI programs to maintain the tempo with rivals. (Gabby Jones / Bloomberg by way of Getty Pictures / File)

The AI Incident Database tracks the “historical past of harms or close to harms” brought on by the proliferation of AI know-how so as “to study from expertise so we are able to forestall or mitigate unhealthy outcomes,” based on the database’s web site. McGregor stated the web site is a response to thinker George Santayana’s aphorism, “Those that can not keep in mind the previous are condemned to repeat it.”

“We should keep in mind previous incidents so we are able to engineer a greater future,” McGregor stated.

The tech professional believes that incidents brought on by AI will doubtless skyrocket this 12 months, based on Newsweek. General, the database has discovered greater than 500 examples of AI incidents from 2003 to this 12 months. Final 12 months ended with 90 incidents, and there have already been 45 incidents this 12 months between January and March. 

Advertisement

CHATGPT AI LISTS JOBS IT CAN DO BETTER THAN HUMANS AS MILLIONS COULD BE PUT OUT OF WORK

“We count on AI incidents to way over double in 2023 and are making ready for a world the place AI incidents are more likely to comply with some model of Moore’s Legislation,” McGregor advised Newsweek, which projected 2023 will see about 180 incidents, based mostly on the information’s present price of incidents.

Moore’s Legislation is an statement from an Intel co-founder that transistors on a circuit would double each two years, making the velocity and functionality of computer systems additionally enhance.

AI photo

As Silicon Valley races to construct highly effective and standard synthetic intelligence programs, troubling “incidents” starting from convincing AI deepfakes, banking fraud, bias and even deaths will enhance this 12 months, a tech professional says. (Reuters / Dado Ruvic / Illustration / File)

The database – which permits customers to look by data such because the 12 months of the incident, the developer behind the AI system or an outline of the incident – reveals that 20 years in the past, there have been 4 experiences of AI incidents. The incidents embody a declare {that a} maternity advert from Goal allegedly predicted a young person was pregnant earlier than her circle of relatives by utilizing an algorithm – although the declare has been referred to as into query.

It additionally consists of two friendly-fire army incidents that left each British and American service members killed through the Iraq Battle.

Advertisement

AI: NEWS OUTLET ADDS COMPUTER-GENERATED BROADCASTER ‘FEDHA’ TO ITS TEAM

The database reveals that varied incidents have elevated since, notably leaping from 37 incidents in 2019 to 78 in 2020, earlier than notching what’s at the moment its highest variety of incidents in 2022 at 90.

“Airplanes cannot be designed to outlive a head-on crash, however that’s the reason we’ve got air site visitors controllers,” McGregor advised Fox Information Digital. “Engineering security requires as many developments in tradition because it requires advances in know-how and that’s what the AI Incident Database is working to assist.”

ChatGPT ai model

McGregor stated he hopes the database will “additionally come to dispel the notion that ChatGPT and related programs are someway god machines which are more and more going rogue.” (Gabby Jones / Bloomberg by way of Getty Pictures / File)

He added that he hopes the database will “additionally come to dispel the notion that ChatGPT and related programs are someway god machines which are more and more going rogue.”

“They’re behaving as we’d count on them to behave based mostly on how they’re engineered,” he stated. 

Advertisement

GOOGLE SCRAMBLES FOR NEW SEARCH ENGINE AS AI CREEPS IN: REPORT

This 12 months alone, the database reported on incidents comparable to a Belgian father allegedly committing suicide after speaking with a chatbot for weeks in regards to the local weather, phony photos of former President Donald Trump getting arrested, and Pope Francis sporting a ritzy white puffer jacket.

“The largest AI incident risk we face proper now’s complacency. By way of every incident we could turn out to be acculturated to the dangers of clever programs somewhat than undertake a ‘by no means once more’ viewpoint,” McGregor stated. 

GERMANY A.I.

The web site of Midjourney, a man-made intelligence able to creating AI artwork, is seen on a smartphone on April 3, 2023, in Berlin. (Thomas Trutschel / Photothek by way of Getty Pictures / File)

The launch of ChatGPT in November served as a watershed second that reverberated throughout the tech trade and catapulted varied firms to compete on constructing their very own AI programs.

OpenAI’s ChatGPT grew to become the fastest-growing person base with 100 million month-to-month energetic customers in January as folks internationally rushed to make use of the chatbot, which ​​simulates human-like conversations based mostly on prompts it’s given.

Advertisement

CYBERATTACKS, AI-HUMAN LOVE ARE MAJOR CHALLENGES OF ARTIFICIAL INTELLIGENCE BOOM, FORMER GOOGLE CHIEF WARNS

Final month, 1000’s of tech leaders, specialists and others signed an open letter calling on all AI labs engaged on tech extra highly effective than ChatGPT to pause for at the very least six months to roll out security laws.

Twitter and Tesla CEO Elon Musk and Apple co-founder Steve Wozniak had been among the many signatories who warned that such laptop intelligence “can pose profound dangers to society and humanity.”

Regardless of the calls to pause, the AI race to create probably the most highly effective system is in full swing, with Google working to overtake its search engine and even create a brand new one which depends on AI, whereas Microsoft has rolled out the “new Bing” search engine described as customers’ “AI-powered copilot for the net.”

Advertisement

Continue Reading
Advertisement
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Technology

Compression-mounted laptop RAM is fast, efficient, and upgradeable

Published

on

Compression-mounted laptop RAM is fast, efficient, and upgradeable

The first laptop with LPCAMM2 memory is finally here. The folks at iFixit got their hands on the Lenovo ThinkPad P1 (Gen 7), which uses LPDDR5X memory in an LPCAMM2 module that you can upgrade or replace by simply unscrewing from the laptop’s motherboard.

That’s a major change from existing LPDDR memory in most modern laptops. This type of memory is typically soldered onto the motherboard, making it much more difficult to upgrade. LPCAMM2 — or Low-Power Compression-Attached Memory Module — offers modularity without losing the power efficiency of LPDDR.

Image: iFixit and Image: iFixit

Traditional DDR RAM uses SODIMM sticks that slot into the motherboard, much like on desktop PCs. As pointed out by iFixit, LPDDR is much more power-efficient, but it needs to be soldered directly to the mainboard, as close to the processor as possible. Up until now, manufacturers have had to choose between upgradeable RAM and power-efficient RAM. LPCAMM2 attempts to split the difference by cozying up to the motherboard with screws, rather than solder. And unlike SODIMM sticks, each LPCAMM2 module is dual-channel, so you only need one, not two.

As shown in the hands-on video posted by iFixit, you can replace the LPCAMM2 in Lenovo’s new ThinkPad P1 by removing the laptop’s backing and then taking out the battery. From there, you can take out the LPCAMM2 by removing three Phillips-head screws.

Advertisement

In addition to Micron, Samsung and ADATA are also getting in on LPCAMM2. That will hopefully mean you’ll be able to upgrade your memory if it comes with a measly 8GB of RAM, but we likely can’t count on every manufacturer to make the shift.

Continue Reading

Technology

Hollywood hijacked: The AI takeover of Tinseltown’s films to fake out Americans

Published

on

Hollywood hijacked: The AI takeover of Tinseltown’s films to fake out Americans

If you’re a fan who counts down the days to upcoming movie releases for your favorite characters, you may also be a big fan of movie trailers. So, you may be quite angry if you believe you’ve stumbled on a secret movie trailer on YouTube that looks like the real thing, only to discover artificial intelligence duped you.

Although people have been creating fake movie trailers for quite a while and releasing them on platforms like YouTube, readily available artificial intelligence apps make the process far easier. A recent fake AI movie trailer for the James Bond movie franchise – that appeared to show Henry Cavill as the new Bond character – became quite popular, generating almost 4 million views in two weeks.

However, the trailer was a fake from KH Studio. Judging from the comments on YouTube, its quality was more than good enough to dupe many viewers who thought they were viewing the real thing. Because it generated so many views, other creators are sure to look to do something similar. 

CLICK TO GET KURT’S FREE CYBERGUY NEWSLETTER WITH SECURITY ALERTS, QUICK VIDEO TIPS, TECH REVIEWS AND EASY HOW-TO’S TO MAKE YOU SMARTER

Movie screen in a movie theater (Kurt “CyberGuy” Knutsson)

Advertisement

Why fake AI movie trailers scare Hollywood

AI technology played a key role in the recent labor talks in Hollywood between studios and the actor’s and writer’s unions. The union members wanted guarantees about how studios would use AI that would protect their jobs and creative rights.

The fake movie trailers highlight some of the biggest fears of the creatives in Hollywood. They fear filmmakers could upload existing material into AI software, allowing the AI to generate new material without needing writers or actors. Perhaps AI software could even reboot original TV shows and movies from decades ago with images of the actors as they looked in their younger days.

Those generating fake AI movie trailers currently don’t have a film studio’s financial resources or computing power. Yet, they are already fooling some people into believing the trailers are real. Without safeguards in place, those studios could potentially use their significant resources in the near future to generate AI content that simply isn’t distinguishable from the real thing.

Hollywood hijacked: The AI takeover of Tinseltown’s films to fake out Americans

Clapperboard used in filmmaking (Kurt “CyberGuy” Knutsson)

MORE: AI’S IMPACT ON HOLLYWOOD AMID ‘BARBENHEIMER’ EPIC FRENZY

The proliferation of fake movie trailers on YouTube

You don’t have to search hard online for advice and step-by-step instructions for creating fake movie trailers. Such instructions include information on mimicking the audio, graphics and video to make the trailer look realistic. They can even help you create a fictional description to accompany the YouTube upload.

Advertisement

Those who create these trailers often are simply looking for ways to show off their creativity – while generating plenty of views for potential monetization.

One creator well-known for these types of movies is Abandoned Films on YouTube. This creator uses a combination of AI and Photoshop to generate fake trailers for modern television shows and films that feature camera technology from the 1950s and 1960s.

Another YouTube creator, Curious Refuge, uses AI to generate movie trailers as if they’re filmed in the style of American filmmaker Wes Anderson.

Hollywood hijacked: The AI takeover of Tinseltown’s films to fake out Americans

Illustration of individual photographic images (Kurt “CyberGuy” Knutsson)

MORE: REAL OR FAKE? AI PRODUCT REVIEWS ARE MAKING IT IMPOSSIBLE TO KNOW THE DIFFERENCE

Protecting creativity and authenticity

As AI advances, the entertainment industry must work closely with technology companies to establish robust safeguards. This may involve developing digital watermarking techniques, improving trailer authentication processes, and ensuring that AI-generated content is clearly labeled. Preserving the integrity of the moviegoing experience will be crucial in this rapidly evolving landscape.

Advertisement

MORE: ARE AI DEEPFAKES THE END OF ACTING AS WE KNOW IT?

Kurt’s key takeaways

No one likes to have the wool pulled over their eyes, especially Hollywood stars. Whether you enjoy the creativity found in fake movie trailers generated with AI or you agree with users on Reddit’s YouTube forum who absolutely hate them, they aren’t going to disappear any time soon. With Hollywood concerned about how AI may affect the entertainment industry, these fake AI movie trailers will surely receive increased attention as they become more prevalent and look more realistic.

Do you believe fake AI movie trailers are something akin to fan fiction? Or do such trailers represent a violation of the rights of the creator of the original works? Let us know by writing us at Cyberguy.com/Contact.

For more of my tech tips and security alerts, subscribe to my free CyberGuy Report Newsletter by heading to Cyberguy.com/Newsletter.

Ask Kurt a question or let us know what stories you’d like us to cover.

Advertisement

Answers to the most asked CyberGuy questions:

Copyright 2024 CyberGuy.com. All rights reserved.

Advertisement
Continue Reading

Technology

Apple iPad event: all the news from Apple’s “Let Loose” reveal

Published

on

Apple iPad event: all the news from Apple’s “Let Loose” reveal

When Apple CEO Tim Cook and a bunch of his deputies take the virtual stage next week to announce new iPads, they’re going to spend a lot of time talking about specs. If the rumors are true, we’re going to get new iPad Pros with OLED screens and thinner bodies, new Airs with faster chips and a correctly placed front camera, and a couple of new accessories. Before they even launch, I feel confident telling you these are the best iPads ever. But after all these years, I still don’t know how to tell you whether you should want an iPad. Or what you’d want to do with it. 

This has been true forever, of course. The iPad is the jack-of-all-trades in Apple’s lineup, a terrific device in many ways that still feels increasingly redundant now that so many people have big phones and long-lasting laptops. Apple seems to have spent the last decade-plus enamored with the idea of the iPad as a shapeshifter — a device that can be exactly what you need at any given time. The company loves that the iPad’s use case is hard to pin down, that it means different things to different people. It’s a fun, good, ambitious idea: The One Gadget To Rule Them All. The way to make that happen, though, is not to upgrade the chips or move the buttons or redesign the rounded corners. It’s to focus less on the iPad itself and more on the things you attach to it.

Continue Reading
Advertisement

Trending