philip lelyveld The world of entertainment technology

OpenAI launches GPT-5, a potential barometer for whether AI hype is justified

...

OpenAI’s presenters also spent time talking about safety improvements to make the chatbot “less deceptive” and stop it from producing harmful responses to “cleverly worded” prompts that could bypass its guardrails ...

At a technical level, GPT-5 shows “modest but significant improvements” on the latest benchmarks, but when compared to GPT-4, it also looks very different and resets OpenAI’s flagship technology in a way that could set the stage for future innovations, said John Thickstun, an assistant professor of computer science at Cornell University. ...

Most recently, OpenAI has said it will turn its for-profit company into a public benefit corporation, which must balance the interests of shareholders and its mission. ...

The inability of a single AI developer to have a “sustained competitive edge” could increasingly force companies to compete on lowering the prices of their AI products, the bank said in a report last month. ...

See the full story here: https://apnews.com/article/gpt5-openai-chatgpt-artificial-intelligence-d12cd2d6310a2515042067b5d3965aa1

Fear of super intelligent AI is driving Harvard and MIT students to drop out

When Alice Blair enrolled in the Massachusetts Institute of Technology as a freshman in 2023, she was excited to take computer science courses and meet other people who cared about making sure artificial intelligence is developed in a way that’s good for humanity.

Now she’s taking a permanent leave of absence, terrified that the emergence of “artificial general intelligence,” a hypothetical AI that can perform a variety of tasks as well as people, could doom the human race. ...

She’s lined up a contract gig as a technical writer at the Center for AI Safety, a nonprofit focused on AI safety research, where she helps with newsletters and research papers. Blair doesn’t plan to head back to MIT. “I predict that my future lies out in the real world,” she said. ...

Other students are terrified of AGI, but less because it could destroy the human race and more because it could wreck their career before it’s even begun. Half of 326 Harvard students surveyed by the school’s undergraduate association and AI safety club were worried about AI’s impact on their job prospects. ...

Even the cofounder of Y Combinator, a startup accelerator known for funding young founders who have dropped out, thinks students should stay in school. “Don’t drop out of college to start or work for a startup,” Paul Graham posted on X in July. “There will be other (and probably better) startup opportunities, but you can’t get your college years back.” ...

See the full story here: https://www.forbes.com.au/news/innovation/agi-fears-is-driving-harvard-and-mit-students-to-drop-out/

Trump Media Is Testing an AI Search Engine Powered by Perplexity

President Donald Trump's media company, Trump Media, is beta-testing an AI search feature on the Truth Social platform, drawing on Perplexity's AI technology.

The new Truth Search AI feature is available now on the web version of Truth Social and will begin public beta testing on its iOS and Android apps at an unnamed future date, the Florida-based company said Wednesday.  ...

UAE state-owned newspaper The National conducted searches using the new product and reported that the AI-generated answers, perhaps unsurprisingly, source conservative-leaning media outlets.

But 404Media was able to get some possibly surprising results. When reporters asked how the American economy is doing, the new search engine said it was "currently facing significant headwinds, with signs of slowdown." ...

See the full story here: https://www.cnet.com/tech/services-and-software/trump-media-is-testing-an-ai-search-engine-powered-by-perplexity/

Breaking Barriers in Film Education: How USC’s School of Cinematic Arts Integrates AI Tools with Runway

...

Tell us about your course and what you were trying to accomplish this semester.
For the last five years, I've been exploring AI and machine learning with an accidental focus on how it could be used in film – it's been five years of nearly daily "aha!" moments. I got lucky to teach AI at USC's School of Cinematic Arts, working with some of the brightest student minds and forward-leaning faculty & administration on Earth. (Though sadly, the rapid development of AI means I never get to use the same syllabus twice!). This semester we focused on integrating acting and directing techniques into AI technology – think "What would Uta Hagen do with AI?" We're very creation-focused – we start with pitch books, turn those images into montage-style trailers, then progress to scenes and directing performances. It's a gentle way to guide them toward completing their dream projects and help them realize nothing's impossible.

...

What excites you most about the future of AI in film education?
What excites me most is seeing students go from having an idea to having a finished piece that gets them meetings in the industry, all within a single semester. The technology removes so many of the traditional gatekeepers and barriers that kept talented people from getting their work out there. It’s not replacing the need for creative vision or storytelling skills – if anything, it makes those more important because more content is coming. The students who understand how to combine traditional filmmaking knowledge with these new tools are going to have incredible opportunities ahead of them.

See the full story here: https://runwayml.com/customers/usc-school-cinematic-arts-partnership

If anyone can make a movie now, what does Hollywood still stand for?

PhilNote: Yves gets extensive screen time in this story!

... Yves Bergquist leads the AI in Media Project at USC’s Entertainment Technology Center, a studio-backed think tank where Hollywood, academia and tech converge. An AI researcher focused on storytelling and cognition, he has spent years helping studios brace for a shift he sees as both inevitable and wrenching. Now, he says, the groundwork is finally being laid.

“We’re seeing very aggressive efforts behind the scenes to get studios ready for AI,” Bergquist says. “They’re building massive knowledge graphs, getting their data ready to be ingested into AI systems and putting governance committees in place to start shaping real policy.”

In an algorithmically accelerated landscape where trends can catch fire and burn out in hours, staying relevant is its own challenge. To help studios keep pace, Bergquist co-founded Corto, an AI startup that describes itself as a “growth genomics engine.” The company, which also works with brands like Unilever, Lego and Coca-Cola, draws on thousands of social and consumer sources, analyzing text, images and video to decode precisely which emotional arcs, characters and aesthetics resonate with which demographics and cultural segments, and why.

“When the game is attention, the weapon is understanding where culture and attention are and where they’re going.” Bergquist says, arguing media ultimately comes down to neuroscience.

Corto’s system breaks stories down into their formal components, such as tone, tempo, character dynamics and visual aesthetics, and benchmarks new projects against its extensive data to highlight, for example, that audiences in one region prefer underdog narratives or that a certain visual trend is emerging globally. Insights like these can help studios tailor marketing strategies, refine storytelling decisions or better assess the potential risk and appeal of new projects.

With ever-richer audience data and advances in AI modeling, Bergquist sees a future where studios can fine-tune stories in subtle ways to suit different viewers. “We might know that this person likes these characters better than those characters,” he says. “So you can deliver something to them that’s slightly different than what you’d deliver to me.”

But adapting won’t be easy, especially for legacy studios weighed down by entrenched workflows, talent relationships, union contracts and layers of legal complexity. “These AI models weren’t built for Hollywood,” Bergquist says. “This is 22nd-century technology being used to solve 21st-century problems inside 19th-century organizational models. So it’s blood, sweat and tears getting them to fit.” ...

See the full story here: https://www.latimes.com/entertainment-arts/movies/story/2025-08-07/hollywood-tomorrow-ai-studios-storytelling-luma-asteria

Citizen Lab director warns cyber industry about US authoritarian descent

Ron Deibert, the director of Citizen Lab, one of the most prominent organizations investigating government spyware abuses, is sounding the alarm to the cybersecurity community and asking them to step up and join the fight against authoritarianism. 

On Wednesday, Deibert will deliver a keynote at the Black Hat cybersecurity conference in Las Vegas, one of the largest gatherings of information security professionals of the year. 

Ahead of his talk, Deibert told TechCrunch that he plans to speak about what he describes as a “descent into a kind of fusion of tech and fascism,” and the role that the Big Tech platforms are playing, and “propelling forward a really frightening type of collective insecurity that isn’t typically addressed by this crowd, this community, as a cybersecurity problem.” ...

Historically, at least in the United States, the cybersecurity industry has put politics — to a certain extent — to the side. More recently, however, politics has fully entered the world of cybersecurity. ...

“If we stay silent when experienced, mission-driven leaders are sidelined or sanctioned, we risk something greater than discomfort; we risk diminishing the very institutions we are here to protect,” Easterly wrote in a post on LinkedIn.  ...

“I think that there comes a point at which you have to recognize that the landscape is changing around you, and the security problems you set out for yourselves are maybe trivial in light of the broader context and the insecurities that are being propelled forward in the absence of proper checks and balances and oversight, which are deteriorating,” said Deibert.

Deibert is also concerned that big companies like Meta, Google, and Apple could take a step back in their efforts to fight against government spyware — sometimes referred to as “commercial” or “mercenary” spyware — by gutting their threat intelligence teams.  ...

Deibert believes there is a “huge market failure when it comes to cybersecurity for global civil society,” a part of the population that generally cannot afford to get help from big security companies that typically serve governments and corporate clients. “This market failure is going to get more acute as supporting institutions evaporate and attacks on civil society amplify,” he said. ...

Deibert is concerned that these threat intelligence teams could be cut or at least reduced, given that the same companies have cut their moderation and safety teams. ...

See the full story here: https://techcrunch.com/2025/08/06/citizen-lab-director-warns-cyber-industry-about-us-authoritarian-descent/

Explainer: What are AI TVs and how do they work?

...

One of the most visible applications of AI in televisions is real-time picture enhancement. AI TVs can recognise the kind of content you're watching—whether it’s an action movie, a cricket match, or a nature documentary—and automatically adjust the brightness, contrast, colour tone, and frame rate accordingly. ...

Sound processing is another area where AI makes a significant impact. AI-powered TVs can identify dialogue versus background noise and adjust the audio output to make speech clearer or create more immersive sound profiles depending on the scene. If you’re watching a thriller in a quiet room at night, the TV might boost dialogue clarity and reduce sudden spikes in volume. ...

Voice recognition and smart assistant integration also fall under the AI umbrella. AI TVs often come with voice-enabled remotes or hands-free controls powered by assistants like Google Assistant or Alexa. ...

Another area of AI integration is personalisation. AI TVs can learn from your viewing habits to recommend content you’re likely to enjoy—across streaming platforms, live TV, and apps. This is similar to how Netflix or YouTube recommends videos, but the AI is embedded into the TV’s operating system itself, creating a much more seamless experience across all platforms. ...

In some newer models of AI TV, AI extends to energy efficiency too. ...

See the full story here: https://www.storyboard18.com/digital/explainer-what-are-ai-tvs-and-how-do-they-work-78243.htm

Top Google Executive Warns AI Is Coming For Your Job: ‘There Is No Middle Class’

A former Google top executive has warned that the emergence of artificial intelligence (AI) could disrupt society as the technology eats away jobs, rendering the middle class non-existent. Mo Gawdat, who left Google X as its chief business officer in 2018, said the 'hell' will begin as early as 2027 as AI eliminates white-collar jobs, with no one spared, including software developers, CEOs, and podcasters.

"The next 15 years will be hell before we get to heaven," Mr Gawdat told British entrepreneur Steven Bartlett on his "Diary of a CEO" podcast. ...

Mr Gawdat said the emergence of AI will trigger 'social unrest' as people still come to terms with losing their livelihoods and sense of purpose, resulting in rising rates of mental health problems, increased loneliness and deepening social divisions. ...

"Unless you're in the top 0.1 per cent, you're a peasant. There is no middle class," he predicted. ...

The warning by Mr Gawdat comes in the backdrop of Geoffrey Hinton, regarded by many as the 'godfather of AI', stating that the technology could soon develop its own language, making it impossible for humans to track the machines. ...

See the full story here: https://www.ndtv.com/offbeat/top-google-executive-warns-ai-is-coming-for-your-job-there-is-no-middle-class-9024185

Bipartisan ‘Block BEARD’ Anti-Piracy Bill Surfaces in Senate

A bipartisan group of U.S. senators has introduced what they’re calling “a discussion draft” of a new anti-piracy bill, the Block Bad Electronic Art and Recording Distributors Act of 2025, known as the Block BEARD Act. Inspired by laws in the UK and Australia, Block BEARD proposes to allow federal courts to order ISPs to block foreign piracy websites that violate U.S. copyright law. It would provide copyright owners who have had their property stolen a new form of relief by having the courts interrupt foreign online piracy operations, preventing them from making unauthorized content available to U.S. households. ...

See the full story here: https://www.etcentric.org/bipartisan-block-beard-anti-piracy-bill-surfaces-in-senate/

ElevenLabs Launches Eleven Music

ElevenLabs has launched Eleven Music, their new AI music generation service, and they say they've done it with a new strategy: don't get sued. While competitors Suno and Udio are in litigation with major labels for allegedly training on copyrighted material, ElevenLabs secured licensing deals with Merlin Network (a digital rights agency for independent labels) and Kobalt Music Group before launching. That's revolutionary thinking in AI circles. ...

On the consumer side, ElevenLabs says it has built safeguards to prevent users from generating songs with specific artist names or copyrighted lyrics. They're essentially creating AI guardrails that acknowledge artistic ownership while still enabling creation. It's like building a copy machine that refuses to photocopy currency. ...

 It's hard to say whether or not you need the biggest artist's works to train a model. It's more important to be able to reference an artist as most people associate a "sound" with a name. You are more likely to prompt, "Give me a song that sounds like Bob Marley" than to prompt, "I need a 72-96 bpm reggae groove with a one-drop pulse, skank guitar, and bubble organ." ...

See the full story here: https://shellypalmer.com/2025/08/elevenlabs-launches-eleven-music/