philip lelyveld The world of entertainment technology

Amazon forms an AI agent-focused lab led by Adept’s co-founder

Amazon says that it’s establishing a new R&D lab in San Francisco, the Amazon AGI SF Lab, to focus on building “foundational” capabilities for AI agents. 

The Amazon AGI SF Lab will be led by David Luan, the co-founder of AI startup Adept, and will seek to build agents that can “take actions in the digital and physical worlds,” and “handle complex workflows” using computers, web browsers and code interpreters. ...

See the full story here: https://techcrunch.com/2024/12/09/amazon-forms-a-new-ai-agent-focused-lab-led-by-adept-co-founder/

Apple hints at radically different AGI view than AI competitors

... In other words, Giannandrea isn’t ruling out Apple being involved in AGI-related breakthroughs, but the end goal isn’t AGI, it’s user-facing products to improve people’s lives. And Apple’s AI head thinks some common AGI optimism is ‘very naive.’ ...

See the full article here: https://9to5mac.com/2024/12/09/apple-hints-at-radically-different-agi-view-than-ai-competitors/

“You can’t have a believable performance without an actor,” Jennifer Howell of Trey Parker and Matt Stone’s Deep Voodoo told a Red Sea Film Fest market panel. But “other jobs, I think, will shift.”

...

Jacquemin mentioned that the public hasn’t really seen any major AI movies from Hollywood released so far. “The interest from studios has gone up a lot over the past year,” Howell said though. “We have met with all the studios. We are doing productions with a handful of them.” Everyone is still in the earlier stages of trying to figure out how to use it though, she concluded. 

Deep Voodo, which two years ago raised $20 million, has used its technology for such content as YouTube series Sassy Justice and a Kendrick Lamar music video. ...

Echoed Jacquemin:... “I think we will see AI effectively democratizing access,” highlighting: “About 20 years ago, when YouTube was created, it ultimately gave birth to an entirely new generation of storytellers.”

See the full story here: https://www.hollywoodreporter.com/movies/movie-news/ai-pro-con-red-sea-souk-wme-superpowers-vision-deepfake-1236080804/

I went to an AI art awards show and it was an eye-opening experience

This is an interesting AI art competition, with mostly China sponsors, mostly European winners, and a global panel of judges. Pictures of the winning submissions at the link.

See the full story here: https://www.creativebloq.com/art/i-went-to-an-ai-art-awards-show-and-it-was-an-eye-opening-experience

Xreal One Smart Glasses Have Entered the Vision Pro Zone, and I Wore Them

Perching a pair of tethered glasses over my own lenses in an office in midtown New York, I see a familiar sight: Floating in front of me is a widescreen Mac monitor. It's just there, hovering in space. I move my head around to take in its size. Ironically, I'm not using a Vision Pro headset to do this. Instead, I'm wearing Xreal's newest glasses, which can cast larger displays and pin them in space, and it works with any device -- laptop, phone, game console -- that can output video via USB-C. ...

How do they do this? The Xreal One and Xreal One Pro glasses -- coming Dec. 9 and early next year, respectively -- have a chip called the X1, custom-made by Xreal, that can keep images fixed in space as you turn your head. These new glasses also have enhanced audio by Bose, a 50- or 57-degree field of view and optional AI-ready snap-in cameras. The Xreal One sounds smarter than Xreal's previous smart glasses. That's exactly the idea. ...

"We made these glasses with 80% of the experience, but at only 20% of their cost in terms of price, weight, everything," Chi says in an exclusive interview. ...

But the best and subtlest part is how displays can gradually move with you or stay fixed in one place. Xreal previously required a separate piece of hardware to get that anchored display to work, the Beam or Beam Pro; now, it's built into the glasses and works with anything -- from PCs and Macs to Steam Decks, tablets and phones. ...

See the full story here: https://www.cnet.com/tech/computing/i-wore-xreal-one-smart-glasses-are-entering-the-vision-pro-zone/

Vitalik Buterin urges Web3 wallets to improve security, privacy

Ethereum co-founder Vitalik Buterin urged Web3 wallet developers to implement features improving privacy and security in a Dec. 3 blog post.

“[A] user only benefits from any decentralization, censorship resistance, security, privacy, or other properties that Ethereum and its applications offer to the extent that the wallet itself also has these properties,” Buterin said in the blog post.

He also proposed ways to ease transfers among Ethereum’s layer-2 (L2) scaling networks, such as Optimism and Arbitrum.

Ethereum hosts dozens of L2 scaling networks, which collectively handle more than $50 billion in total value locked (TVL), according to data from L2Beat. ...

On Nov. 26, a United States court ruled in favor of Tornado Cash, saying the US Treasury Department overstepped in sanctioning the privacy protocol. This potentially paves the way for more onchain privacy features.

Buterin also recommended integrating multisignature authorization — where several signers must authorize transactions — into all Web3 wallets for added security. ...

See the full story here: https://cointelegraph.com/news/vitalik-buterin-web3-wallets-improve-security-privacy

USC student embracing AI for filmmaking and art

... Holly Willis is the chair of USC’s Media Arts and Practice Program, as well as the co-director of the AI for media and storytelling studio. She says on top of saving precious time in putting a film together, AI is also helps keep down cost when it comes shooting, and producing portions of films and art. ...

See the full story here: https://spectrumnews1.com/wi/madison/entertainment/2024/11/28/usc-ai-storytelling

Licking this “lollipop” will let you taste virtual flavors

...

A team of Hong Kong scientists has now developed a handheld user interface shaped like a lollipop capable of re-creating several different flavors in a virtual environment, according to a new paperpublished in the Proceedings of the National Academy of Sciences (PNAS).

It's well established that human taste consists of sweet, salty, sour, bitter, and umami—five basic flavors induced by chemical stimulation of the tongue and, to a lesser extent, in parts of the pharynx, larynx, and epiglottis. Recreating those sensations in VR has resulted in a handful of attempts at a flavor user interface, relying on such mechanisms as chemical, thermal, and electrical stimulation, as well as iontophoresis. ...

So Yiming Liu of City University of Hong Kong and co-authors opted to work with iontophoresis, in which stable taste feedback is achieved by using ions flowing through biologically safe hydrogels to transport flavor chemicals. This method is safe, requires low power consumption, allows for precise taste feedback, and offers a more natural human-machine interface. Liu et al. improved on recent advances in this area by developing their portable lollipop-shaped user interface device, which also improves flavor quality and consistency. ...

 There were nine taste-generating channels filled with flavored hydrogels. The gels were made out of agarose mixed with a bit of mineral water and specific flavor essences: sugar, salt, citric acid, cherry, milk, green tea, passion fruit, durian, and grapefruit. ...

Flavors are generated via a current flowing through a target gel, delivering flavor chemicals to the outside of the lollipop. ...

See the full story here: https://arstechnica.com/science/2024/11/licking-this-lollipop-will-let-you-taste-virtual-flavors/

AI In Entertainment: 19 Practical And Ethical Challenges

...

8. ‘Filter Bubbles’

A key challenge is AI-driven content personalization, which can create “filter bubbles” and limit the diversity of content recommendations for viewers. While AI optimizes engagement, it may also result in repetitiveness and reduced exposure to new genres. To counter this, the industry should leverage AI to enhance diversity by expanding recommendation algorithms to prioritize novelty alongside user preferences. - Santhosh Vijayabaskar

...

16. The Erosion Of Cultural Diversity

One challenge is the erosion of cultural diversity as AI-generated content becomes prevalent. AI might prioritize patterns lacking cultural nuance, leading to a homogenized entertainment landscape. Addressing this requires integrating diverse datasets and collaborating with creators from varied backgrounds to ensure AI tools amplify, not suppress, cultural richness. - Nicola Sfondrini, PWC

...

See the full story here: https://www.forbes.com/councils/forbestechcouncil/2024/11/26/ai-in-entertainment-19-practical-and-ethical-challenges/

Jerry Garcia’s AI Voice Debuts in ElevenLabs’ Audiobook App

ElevenLabs is pushing the boundaries of AI narration by introducing Jerry Garcia’s iconic voice to its Eleven Reader audiobook app. Fans of the legendary Grateful Dead frontman can now hear his distinctive tones narrating a wide range of content, from audiobooks and articles to poetry and PDFs. Dubbed the “Iconic Listening Experience,” this feature extends Garcia’s AI narrator voice to 32 languages, making his legacy accessible to a global audience.

Garcia joins an illustrious lineup of celebrity voice AI narrators on the ElevenLabs platform, including Maya Angelou, John Wayne, and James Dean. This innovation isn’t limited to nostalgic entertainment—it’s also reshaping how we interact with information and creativity. ...

...the Jerry Garcia Foundation, co-founded by Garcia’s daughter, Keelin Garcia, is contemplating using her father’s AI model to narrate documentaries and audio art exhibits. ...

See the full story here: https://www.eweek.com/news/jerry-garcia-ai-voice-eleven-labs/