MAROKO133 Eksklusif ai: Disney+ Will Allow Users to Generate Their Own “Frozen 3” Using AI

📌 MAROKO133 Hot ai: Disney+ Will Allow Users to Generate Their Own “Frozen 3” Usin

Disney is planning to flood its streaming service, Disney+, with user-generated AI slop.

During the company’s recent earnings call, Disney CEO Bob Iger said that the streaming service is “in the midst of rolling out the biggest and the most significant changes” since its inception six years ago, as quoted by The Hollywood Reporter.

Specifically, Iger was referring to what sounds a bit like OpenAI’s Sora: a service that allows users to generate AI content on the service. Why wait for the long-awaited sequel to Disney’s “Frozen 2” when you can just cook up your own AI-generated take?

“The other thing that we’re really excited about, that AI is going to give us the ability to do, is to provide users of Disney+ with a much more engaged experience, including the ability for them to create user-generated content and to consume user-generated content — mostly short-form — from others,” Iger said.

While plenty of questions remain about how exactly Disney is hoping to roll out the new feature, OpenAI’s foray into user-generated AI slop should probably serve as a cautionary tale. The ChatGPT maker has attracted plenty of negative attention with its Sora app, from glaring instances of copyright infringement to users generating highly problematic content.

Besides, as the Wall Street Journal reported in August, Disney had already scrapped several AI projects over legal concerns that using AI to clone actors could draw anger and retaliation from human performers and trade unions.

Disney’s latest gambit is an especially surprising development considering Disney’s extensive efforts to protect its own intellectual property. Disney has fought to keep its own intellectual property from appearing in Sora, as Reuters reported in September. Disney also sent a cease and desist letter to Character.AI, accusing it of using its copyrighted characters without permission.

Iger claims that the company had “productive conversations” with yet-unnamed AI companies to reach an agreement that would “reflect our need to protect the IP,” as per THR.

Regardless, Disney will likely also have to double down on content moderation to ensure that the platform remains family-friendly — which is far easier said than done, as previous failures to implement meaningful AI guardrails, including effective age restrictions, have demonstrated.

Iger also hinted at the possibility of integrating a “number of game-like features into Disney+,” based on its agreement with video game developer Epic Games.

“The opportunity here, we think, is enormous in terms of increasing our engagement with Disney fans across the world,” Iger gushed during this week’s earnings call.

But whether the reality of launching a short-form AI-generated content feature will live up to those lofty promises remains anything but certain. Given the mayhem that has unfolded following OpenAI’s launch of its Sora app, Disney will likely have its work cut out to ensure that such a feature won’t immediately plunge its streaming service into chaos.

It certainly would be far from the first time Disney’s IP has been abused with the help of AI on the web. A quick search on Sora reveals a litany of photorealistic renditions of Disney movie clips, among other videos featuring the company’s widely known characters.

More on Disney: Disney’s Secret Experiments With AI Have Reportedly Been a Comical Disaster

The post Disney+ Will Allow Users to Generate Their Own “Frozen 3” Using AI appeared first on Futurism.

🔗 Sumber: futurism.com

📌 MAROKO133 Breaking ai: Weibo's new open source AI model VibeThinker-1.5B ou

Another day in late 2025, another impressive result from a Chinese company in open source artificial intelligence.

Chinese social networking company Weibo's AI division recently released its open source VibeThinker-1.5B—a 1.5 billion parameter large language model (LLM) that is a fine-tuned variant of rival Chinese tech firm Alibaba's Qwen2.5-Math-1.5B.

It's available now for free download and usage by researchers and enterprise developers—even for commercial purposes—under a permissive MIT License on Hugging Face, GitHub and ModelScope, with a technical report on open access science publishing site arxiv.org.

And yet, despite its compact size, VibeThinker-1.5B achieves benchmark-topping reasoning performance on math and code tasks, rivaling or surpassing models hundreds of times its size, even outperforming Chinese rival DeepSeek's famed R1 that went viral at the start of this year—a 671-billion parameter model—on formal reasoning benchmark.

It further eclipses Mistral AI's Magistral Medium and holds its own against Anthropic's Claude Opus 4 and OpenAI's gpt-oss-20B Medium, all while requiring a fraction of the infrastructure and investment.

It also does so having been post-trained on a budget of merely $7800 USD for compute resources (3900 GPU hours on Nvidia H800s) — far less than the tens, or even hundreds, of thousands of dollars typically required to fine-tune models of similar or larger scale.

Recall this is not the total cost of the model's development, however: LLMs are trained in stages. First comes pre-training, when the model learns basic language structure and general knowledge by predicting the next word across enormous amounts of text from the internet, books, and articles. This gives it fluency but not much sense of how to follow instructions or hold a conversation

Post-training comes next, using much smaller, higher-quality datasets—typically collections of example questions, prompts, and expert-written answers—to teach the model how to respond helpfully, reason through problems, and align with human expectations. Still, Weibo's post-training cost effectiveness on VibeThinker-1.5B is noteworthy and should be commended.

The open-source release upends assumptions about parameter scale, compute intensity, and the minimum viable size for high-performance LLMs.

A Different Training Approach: Spectrum-to-Signal

VibeThinker-1.5B owes its performance not to scale, but to the training framework behind it: the Spectrum-to-Signal Principle (SSP).

Instead of optimizing a model purely for single-answer correctness (Pass@1), the SSP framework decouples supervised fine-tuning (SFT) and reinforcement learning (RL) into two distinct phases with different goals:

• SFT (“Spectrum Phase”): The model is trained to maximize diversity across potential correct answers, improving its Pass@K score. This builds a wide range of plausible solution paths.

• RL (“Signal Phase”): A second-stage reinforcement learning system (called MaxEnt-Guided Policy Optimization, or MGPO) is used to identify and amplify the most correct paths from this diverse solution pool. MGPO prioritizes problems where the model is most uncertain, using entropy-based weighting to focus learning.

The authors argue this separation allows small models to explore reasoning space more effectively—achieving signal amplification without relying on massive parameter counts.

VibeThinker-1.5B makes a compelling case that the industry’s reliance on parameter scaling as the only route to better reasoning performance may be outdated.

By adopting a diversity-first training pipeline, WeiboAI has shown that smaller, more accessible models can match and even outperform billion-dollar systems in logic-heavy tasks.

The low resource footprint is among the most significant aspects of VibeThinker-1.5B. At under $8,000, the post-training cost is 30–60x lower than models like DeepSeek R1 and MiniMax-M1, which cost between $294K and $535K to train.

Performance Across Domains

Despite its small size, VibeThinker-1.5B delivers cross-domain reasoning that outpaces many larger open-source and commercial models:

VibeThinker was benchmarked against both reasoning-centric models (Magistral, Claude, OpenAI o3-mini) and non-reasoning LLMs (GPT-4.1, Kimi K2, DeepSeek V3). Across structured reasoning benchmarks, the model consistently outperformed non-reasoning models, regardless of size:

• On AIME24 (math), it beat Kimi K2 (1.09T) by over 10 points (80.3 vs. 69.6).

• On LiveCodeBench v6, it surpassed Claude Opus 4 (51.1 vs. 47.4).

• On GPQA, it scored below GPT-4.1 and Claude, but still doubled its base model (from 16.4 to 46.7).

This supports the authors’ claim that size is not the only path to reasoning capability—with proper training design, smaller models can reach or even exceed the performance of far larger systems in targeted tasks.

Notably, it achieves parity with models hundreds of times larger on math and code, though it lags behind in general knowledge reasoning (GPQA), where larger models maintain an edge.

This suggests a potential specialization trade-off: while VibeThinker excels at structured logical tasks, it has less capacity for wide-ranging encyclopedic recall, a known limitation of smaller architectures.

Guidance for Enterprise Adoption

The release includes recommended inference settings (temperature = 0.6, top_p = 0.95, max tokens = 40960).

The model is small enough to be deployed on edge devices, including mobile phones and vehicle-embedded systems, while inference costs are estimated to be 20–70x cheaper than with large models.

This positions VibeThinker-1.5B not just as a research achievement, but as a potential foundation for cost-efficient, locally deployable reasoning systems.

Weibo’s Strategy and Market Position

Weibo, launched by Sina Corporation in 2009, remains a cornerstone of China’s social media ecosystem. Often described as China’s version of X (formerly Twitter), the platform blends microblogging, multimedia content, and trending-topic features with a regulatory environment shaped by tight government oversight.

Despite counting 600 million monthly active users (more than twice that of X), investors are not optimistic about its advertising revenue growth potential in the near term, and Weibo is navigating intensifying competition from video-first platforms like Douyin, which are drawing younger users and increasin…

Konten dipersingkat otomatis.

🔗 Sumber: venturebeat.com