YouTube Announces Veo2 Integration for Shorts Creators

⚡️ Headlines

🤖 AI

Apple partners with Baidu to enhance iPhone AI features in China – The collaboration focuses on integrating advanced AI capabilities tailored for the Chinese market. [The Information]

Baidu announces Ernie AI model will be available for free use starting April – The move aims to promote the broader adoption of Baidu’s generative AI across industries. [Reuters]

Apptronik secures $350M to mass-produce AI-powered humanoid robots – Funding will support scaling production to meet growing global demand. [GlobeNewswire]

Elon Musk’s offer letter to buy OpenAI reveals five key negotiation details – Musk's vision for OpenAI centers on transparency and accelerated development. [TechCrunch]

DeepSeek is India’s push for AI sovereignty amid global AI race – India seeks to strengthen its domestic AI ecosystem through strategic initiatives. [Bloomberg Opinion]

UK’s AI Safety body rebrands to AI Security Institute, signs MOU with Anthropic – The shift reflects a focus on national security in AI development. [TechCrunch]

🦾 Emerging Tech

Wall Street giants like BNY and Citi ramp up involvement in crypto – Financial institutions deepen their exposure despite ongoing regulatory challenges. [The Information]

Coinbase sees revenue surge driven by U.S. election-fueled crypto rally – The post-election rally boosts trading volume and profits. [The Information]

🤳 Social Media

TikTok and Luminate unveil 2025 Music Impact Report – The report highlights TikTok’s growing influence on global music discovery and consumption. [TikTok Newsroom]

Giphy reveals Valentine’s Day search trends in latest report – Popular searches reflect users’ romantic and humorous moods around the holiday. [Giphy]

TikTok ban temporarily lifted as it returns to iOS App Store and Google Play – The move follows legal battles over privacy and data security. [Wired]

Meta faces scrutiny over AI’s use of personal DM data – The company’s AI training methods raise privacy concerns. [Social Media Today]

⚖ Legal

Cohere faces copyright lawsuit from publishers over AI training data – Publishers claim Cohere used protected content without proper authorization. [TechCrunch]

🎱 Random

Arm plans to launch its first custom chip this year, with Meta as a key customer – The chip will power AI workloads, expanding Arm’s influence in hardware development. [TechCrunch]

🔌 Plug-Into-This

Imagine it, create it: Veo 2 is coming to YouTube Shorts

YouTube Shorts introduces Veo 2, powered by Google DeepMind’s latest AI video generation technology. Users can now create detailed and realistic video clips using text prompts, which can be added to Shorts content. Veo 2 enhances the Dream Screen feature by generating high-quality, customizable videos quickly.

• Veo 2 supports standalone video clip generation via text prompts.

• Dream Screen now integrates improved physics and movement realism.

• Users can customize videos with specific styles, lenses, or cinematic effects.

• AI-generated videos include watermarks for transparency.

• Features are set to be available in the U.S., Canada, Australia, and New Zealand.

🎥 Integrating Veo 2 into Shorts for content creators signals Google’s intent to compete by offering it’s advanced AI products within existing workflows, driving adoption by pushing them out to established users. Shorts is hardly TikTok, but could be well positioned to capture wayward creators if TikTok does ultimately get banned (or gutted during a sale).

Google's Hassabis warns of AI race's hazards

In a recent interview, Google DeepMind CEO Demis Hassabis expressed concerns that the intensifying competition in artificial intelligence development could compromise safety protocols. He emphasized the need for international collaboration to establish norms and standards, cautioning that a race-like environment makes it challenging to ensure the technology remains safe.

• Hassabis highlighted that the rapid pace of AI advancement might lead organizations to prioritize speed over thorough safety measures.

• He advocates for a global framework, similar to the Intergovernmental Panel on Climate Change, to oversee AI development and mitigate potential risks.

• The DeepMind chief has previously compared the risks associated with uncontrolled AI to significant global challenges like climate change and pandemics.

• Despite acknowledging the potential dangers, Hassabis remains optimistic about AI's capacity to drive scientific breakthroughs and address complex problems.

• He has called for increased investment in AI safety research to develop robust mechanisms that can keep advanced AI systems aligned with human values.

🤷 The call for international cooperation against the backdrop of the more or less failed AI action summit (at least in terms of agreeing on safety regulations) is probably nothing more than performative hedging from Google’s chief. On a core level, they’re likely pursuing just as much advancement as the rest — but having enjoyed a couple decades of monopoly in the search engine market, might as well try to make one in AI too, right? Regulations are one way to achieve that, if given enough sway.

Anthropic's Next Major AI Model Could Arrive Within Weeks

Anthropic is preparing to launch a new AI model described as a "hybrid" capable of switching between "deep reasoning" and fast responses. This development aims to enhance the model's versatility in handling complex tasks efficiently.

• The upcoming model is expected to improve upon previous iterations by balancing thorough analytical capabilities with the ability to provide quick answers when needed.

• This dual-mode functionality is designed to cater to diverse user requirements, allowing the AI to adapt its response strategy based on the complexity of the query.

• The "deep reasoning" mode will enable the AI to perform in-depth analysis and generate comprehensive responses for complex problems.

• Conversely, the fast response mode is intended for more straightforward queries, delivering prompt answers without unnecessary processing delays.

• This approach reflects Anthropic's commitment to advancing AI technology that is both adaptable and efficient, meeting a wide range of user needs.

🚀 The anticipated release of this hybrid AI model signals Anthropic’s commitment to competing with OpenAI and Google, perhaps showing a strategy of combining multiple capacities into one — mirroring OpenAI’s recent announcement re: GPT-5.

 🆕 Updates

📽️ Daily Demo

🗣️ Discourse