OpenAI’s Film Goes to Cannes, Google Safety Backlash, US Schools Prioritize AI

A peer-reviewed study from Charles Darwin University says AI is eroding human dignity worldwide as regulation lags behind. Lead author Dr Maria Randazzo reports current laws overlook privacy, anti-discrimination, user autonomy, and intellectual property rights.

The research cites the “black box” opacity of machine-learning models, noting users cannot trace or contest harmful decisions. It contrasts the market-centric U.S., state-centric Chinese, and human-centric EU approaches, concluding none alone adequately protect citizens.

Randazzo views the EU model as the strongest template yet still insufficient without coordinated global adoption. She warns the problem is “only going to get worse” and will deepen biases and weaken democratic values if unaddressed.

OpenAI is supplying its generative tools and computing power to produce “Critterz,” a feature-length animated film conceived by in-house creative specialist Chad Nelson. The project expands a 2023 OpenAI-funded short into a full theatrical release and aims to debut at Cannes in May.

Vertigo Films and Federation Studios are financing the movie on a budget under $30 million, a fraction of typical Hollywood animation costs. The team plans to finish the film in nine months and release it in theaters worldwide next year.

The nine-month timeline contrasts with the three-to-four-year schedules common in animation. Its sub-$30 million budget sits far below Disney’s reported $200 million for Tangled and $130 million for Tarzan, underscoring the scale of potential savings highlighted in the article.

When the former Zillow exec who sold his first company for $120m starts a new venture, people notice. Thousands of people; 10,000+ regular investors have already joined the early investors behind Uber, eBay, and Venmo as early-stage Pacaso investors.

Disrupting the real estate industry once again, Pacaso’s streamlined platform offers co-ownership of premier properties, revamping the $1.3t vacation home market. And it works.

By handing keys to 2,000+ happy homeowners, Pacaso has already made $110M+ in gross profits in less than 5 years. They even recently reserved the Nasdaq ticker PCSO. And you can become a Pacaso shareholder before the investment opportunity ends September 18.

The State Educational Technology Directors’ Association survey finds 26% of state technology officials now call AI their top priority, pushing cybersecurity to second place. More than 75 leaders across 47 states participated, marking the first time AI leads the annual ranking.

Forty percent of respondents say their state has issued AI guidance, a jump from just 2% two years ago. North Carolina, Utah, and Wyoming are appointing AI leads, brokering statewide tool deals, and bankrolling teacher training programs.

SETDA’s executive director notes “the rubber’s hitting the road” as focus moves from guidance to classroom integration. Officials cite urgent needs around student AI literacy, educator professional development, and policies on cheating and proper use.

Kids-safety nonprofit Common Sense Media labels Google’s Gemini AI offerings for children and teens “High Risk” in a newly published safety assessment. The review highlights that the chatbot still delivers content about sex, drugs, alcohol, and mental health advice to minors.

Common Sense says the Under 13 and Teen experiences run the adult Gemini model with only add-on filters, arguing true kid safety needs ground-up design. The report arrives as leaks suggest Apple may adopt Gemini for Siri, potentially widening youth exposure.

Google disputes the findings yet concedes some responses “weren’t working as intended” and says it has added further safeguards. The rating places Gemini on the same risk tier as Perplexity, above ChatGPT’s “moderate” label and below the “unacceptable” score given to Meta AI and Character.AI.

LunaLock ransomware hackers breached the Artists&Clients commission site, encrypted all files, and demanded at least $50,000 in Bitcoin or Monero. They warn that if payment isn’t made, stolen artwork and personal data will be leaked and sold to AI companies for model training.

Artists&Clients markets itself as an artist-only marketplace that forbids AI involvement, making the threat particularly pointed. The site has been offline since the August 30 attack, and management has issued no statement while users worry about exposed messages, artwork, and payment information.

Security researcher Tammy Harper says this is the first known ransomware case that explicitly leverages AI training as a coercion tool. Harper notes the approach targets creators who already oppose their work being used for AI training without compensation.