Nicolas Cage Rejects Amazon AI Waiver, Parents Ban Social Media, Microsoft Challenges OpenAI, Elon Musk Pushes Credit Cards Into Grok

SPEAKER_1 (0:05)

Welcome to Digimasters Shorts.

SPEAKER_2 (0:07)

We are your hosts, Adam Nagus and Carly Wilson, delivering the latest scoop from the digital realm.

Nicolas Cage says he refused to sign an artificial intelligence waiver for Amazon's Spider-Noir. Speaking at New York Comic Con, he linked his decision to Amazon's $50 billion investment in OpenAI. Cage said Amazon's support for AI could mean he will not work with the studio again. Prime Video cancelled Spider-Noir after one season, despite the series earning 11 Emmy nominations. The show drew 2.6 billion viewing minutes during its first six weeks, and marked Cage's first regular television role. Cage remains connected to Amazon through the upcoming John Madden biopic Madden.

He also said he wants to work with younger filmmakers emerging from YouTube, including Backroom's director Kane Parsons. Cage added that he may appear in a future season of HBO's True Detective.

SPEAKER_1 (0:58)

Parents are increasingly restricting their children's access to social media and artificial intelligence amid growing concerns about safety, addiction and privacy. Social media companies have faced accusations of designing addictive products for young users, while OpenAI and other AI firms have been criticized over insufficient chatbot safeguards. Although both industries deny the claims and have introduced measures such as parent-linked teen accounts, many families remain unconvinced. A survey of parents found that most respondents no longer wanted their children using social media, while many also tried to limit their use of AI. Some parents allow messaging services for real-world connections, but prohibit social platforms, while others permit supervised chatbot use and teach children to question AI-generated information.

Environmental concerns about AI's energy and water consumption have also led some families to avoid these tools when alternatives are available. Parents working in technology say children should understand AI, but argue they should not become unwitting test subjects for products whose effects on learning and well-being remain uncertain. They also criticize schools and authorities for introducing digital tools without sufficient parental involvement, digital literacy education or critical thinking guidance. Many families describe managing screen limits as stressful, with children finding ways around restrictions and digital media displacing reading, outdoor activities and face-to-face relationships.

Some parents now support collective efforts to lay smartphones and social media, arguing that stronger regulation and family cooperation are necessary to protect children.

Microsoft has released Microsoft Decision One, a small model designed to choose among fixed answers rather than generate text. Built on Alibaba's open weight, Quen 3.59b, it provides calibrated probabilities for yes or no questions, multiple choice tasks, ratings, and AI response evaluations. Microsoft says it achieved the highest accuracy across 36 benchmarks covering nearly 150,000 questions. The company also reported that it was 2.5 times faster than the runner-up, and 35 times faster than OpenAI's GPT-6 Sol in its tests. The model changed its decisions in just 1.3 percent of tests, when requests were reworded or reordered. Microsoft has tested it for safety risks including harmful content, jailbreaks, and prompt injection. Xbox Research used the model to classify more than 10,000 pieces of player feedback, while Microsoft's Co-Pilot team reported significant speed and cost advantages. The model costs 0.042 dollars per million input tokens with no charge for output tokens, and is available through Microsoft Foundry and OpenRouter. Microsoft plans to re-base the model on its own MI systems and OpenAI models, as competition intensifies among more than 100 decision-focused AI models.

SPEAKER_2 (3:45)

Elon Musk sparked criticism after suggesting that X users upload photos of their credit cards to Grok for online shopping. He claimed the AI could use the information to find the best deals and complete purchases. Security warnings from XAI advise users not to include personal information in prompts. Grok's terms also state that user data and credentials may be stored in the cloud and reviewed by humans during financial transactions. When asked whether sharing a credit card image was safe, Grok advised against uploading an actual card. The conflicting guidance prompted memes, jokes and requests for Musk's own credit card details. Some users even jokingly suggested sharing their DNA with the chatbot. Security experts continue to warn that credit card images should never be posted in AI chats or online.

SPEAKER_3 (4:27)

Thank you for listening to today's AI and tech news podcast summary. Please do leave us a comment and for additional feedback, please email us at podcast at digimasters.co.uk.

You can now follow us on Instagram and threads by searching for Digimasters Shorts or search for Digimasters on LinkedIn. Be sure to tune in tomorrow and don't forget to follow or subscribe.