Xiaoliu BOT
中文EN

X Platform April 25 AI Brief: Grok Imagine Upgrades Lip Sync, AI Video Workflow Simplified, DeepSeek Valuation Draws Attention

Scan timeline count: 360 | Hits bloggers: 23 | Total hit tweets: 192 | Weighted tweet score: 145.65 | Original tweets: 75 | RT tweets: 44 | Fetch attempts: 2 | Boundary coverage status: Full coverage

Controversy Storm Over SPLC & “Anti-Hate” Organizations

Multiple bloggers mentioned that the Southern Poverty Law Center (SPLC) has recently sparked widespread controversy due to a grand jury indictment, with various sources citing that the SPLC historically funded the KKK, revived neo-Nazi organizations, and used “manufacturing hate” to attract donations. The core argument is that it is precisely the “elimination of hate” that would end the existence of these organizations, so they instead amplify hate.

> The SPLC invested $270,000 in the Charlottesville incident but received an additional $81 million in donations, an astonishing return on investment, which is a manifestation of the incentive mechanism. The SPLC changed the definition of “white supremacy” and applied it to everyone—supply-side issues were thus “solved.” The ADL listed Turning Point USA, founded by Charlie Kirk, as an “extremist threat” and “hate group,” placing it alongside real terrorists. The NAACP is a partner of the SPLC, and both are continuously inciting hatred.

Source:

Major Grok Imagine Update: Significant Lip Sync and Audio Quality Enhancement

Multiple bloggers mentioned that Grok Imagine has just released a new version, significantly improving lip sync and audio quality in image-to-video generation. Grok Voice ranked first in multiple real-world voice AI benchmarks, with an overall score of 67.3% and 62.3% in retail scenarios, also leading in AI-specific tests. Grok’s performance in Hebrew and other languages also surpassed Gemini.

Source:

SpaceX & Starship: Falcon Heavy Footage and Iteration Speed Elicit Awe

Multiple bloggers mentioned that SpaceX released footage of the Falcon Heavy inside the LC-39A launch pad in Florida, with the heavy-lift rocket standing quietly in the hangar awaiting launch. Meanwhile, the iteration speed of Starship since its first flight three years ago has been incredible, with SpaceX progressing at near-light speed. Starlink has already replaced its customer service center with AI, significantly boosting operational efficiency.

Source:

Hot AI Video Generation Tools: GPT Image 2 + Seedance 2.0 Combined Workflow

Multiple bloggers mentioned that cellinlab shared a complete workflow for generating a zero-editing AI-generated K-pop girl group MV: first using GPT Image 2 to generate 12-frame storyboards, then feeding them into ByteDance’s official Seedance 2.0 to produce the full MV, all within 10 minutes. Some bloggers also tested a solution using 48 frames for smoother transitions and longer duration. In terms of character aesthetics, some noted that GPT Image 2 still lags behind Banana Pro.

Source:

DeepSeek V4 Open Source and Vast Valuation Gap Between Chinese and Foreign AI

One blogger mentioned that DeepSeek’s latest valuation has reached $40+ billion, far exceeding the previously rumored $10 billion. Zhipu is valued at $50 billion, MiniMax at $30 billion, and Kimi, which is about to list on the Hong Kong stock market, at approximately $20 billion. Compared to US companies OpenAI at $850 billion and Anthropic at $1 trillion, the gap is 20-50 times, far exceeding exchange rate factors. ColaOS has integrated DeepSeek V4 immediately.

Source:

Exploration of AI-Generated Anime-Style Illustrations and Commercial Viability

Multiple bloggers mentioned that creators are using AI tools for commercial-grade illustration creation, with 10 styles of national manga illustrations finalized in 1 minute, sparking discussions about “whether it can be commercialized.” Some also tested AI’s understanding boundaries regarding character identity, finding that models can generate hallucinations inconsistent with expectations for specific character settings. The gap between professional camera mode and smartphone photography in AI visual understanding is evident.

Source:

Philosophical Reflections on the Meaning of Human Existence in the Context of Approaching AGI

One blogger mentioned having a conversation with AI about “what remains for humans as AGI approaches,” introducing core insights from Heidegger’s “Being and Time”: humans have no preset purpose, unlike a hammer whose meaning is locked from the moment it leaves the factory; humans are thrown into the world and must define themselves; most people’s lives are not in their own hands, and when the pressures of finite time and mortality bear down, they begin to make every choice seriously.

Source: