Good morning, it’s Wednesday. In today’s edition, Google's AI chief is laying out the game plan for 2025, and it’s all about AI agents that don’t just assist you — they act for you. Think fewer prompts, more autonomy, and an AI ecosystem woven seamlessly into your life.
Meanwhile, OpenAI pushes to reduce nonprofit control, while the UK’s AI copyright opt-out plan faces backlash from creators. Let’s dive in!
🗞️ YOUR DAILY ROLLUP
Top Stories of the Day
🤖 NVIDIA Unveils Budget-Friendly AI Supercomputer for Creators
NVIDIA has introduced the Jetson Orin Nano Super Developer Kit, a compact generative AI supercomputer delivering a 1.7x performance boost for just $249. Targeted at developers, students, and hobbyists, it offers 67 INT8 TOPS, high memory bandwidth, and robust AI model support for robotics, computer vision, and more. Integrated with NVIDIA’s AI ecosystem, it's an affordable powerhouse for prototyping and AI innovation.
🔍 ChatGPT Search Expands with Key Mobile Upgrades
OpenAI is rolling out its AI-powered search engine to all ChatGPT users, including free-tier accounts, during its “12 Days of Ship-mas” campaign. The update brings mobile-friendly enhancements, such as location-based results with ratings, hours, images, and maps, alongside quicker access to relevant websites. Paid users also get Advanced Voice Mode with real-time web info, marking OpenAI's challenge to traditional search engines.
⚖️ UK’s AI Copyright Opt-Out Plan Faces Backlash
The UK government proposes allowing AI companies to use copyrighted works for training unless creators opt out. Critics, including artists and publishers, argue this could harm smaller creators, with notable opposition from figures like Thom Yorke and Julianne Moore. While the plan seeks to balance innovation and creator rights, campaigners demand stricter rules, especially against AI replicating celebrity voices and likenesses.
🚀 OpenAI's o1 Model API Launches with Upgrades
OpenAI has fully launched its o1 model API, offering faster, smarter features like guided messaging, visual input support, and improved accuracy—boosted by 35% with 60% fewer tokens. Real-time voice APIs now include WebRTC support for easier integration and reduced audio costs. Developers can refine outputs with “direct preference optimization” and access new Go and Java SDKs, while the advanced o1 Pro model remains on the horizon.
🔮 2025 PREDICTIONS
Google Executive Predicts AI Agents Will Redefine Automation in 2025
The Recap: In 2025, Google sees the next phase of AI evolving toward "agentic capabilities," where AI assistants will become more autonomous, personalized, and seamlessly integrated into daily life. Sissie Hsiao, Google’s vice president of Gemini, highlights how AI will act on users’ behalf across platforms, marking a significant leap beyond simple conveniences.
Google’s Gemini AI is already assisting with complex tasks like debugging code and practicing for job interviews using features like Gemini Live.
In 2025, Google aims to develop AI agents that can autonomously complete tasks, remember user preferences, and act across Google services, third-party platforms, and the web.
Google’s Deep Research feature in Gemini Advanced is an early example of AI agents, simplifying complex topics into digestible reports.
Gemini Live, launched this year, supports natural, conversational voice interactions with interruptions, offering 10 distinct voices and support for 12 languages.
Multimodal AI, capable of processing speech, images, and video, will be a major focus in the next phase of AI development.
Google positions AI as essential for businesses to stay competitive, comparing the AI ecosystem to the evolution of smartphones, where apps and integrations are crucial.
Forward Future Takeaways:
AI in 2025 will go beyond reactive tools, transitioning into proactive, agentic systems capable of managing tasks autonomously and tailoring experiences to users’ needs. For consumers and businesses, this shift will redefine expectations for AI integration, making it as essential as smartphones and internet services. Companies that embrace AI as a foundational tool, rather than an add-on, will lead the next wave of innovation. → Read the full article here.
👾 FORWARD FUTURE ORIGINAL
The Philosophy of AI | Part II
What is philosophy? What is existentialism?
Philosophy is one of the oldest disciplines in human history. It arose from the need to ask fundamental questions about the world and our existence: What is truth? What is justice? What constitutes a good life? Unlike empirical sciences, which gain knowledge through experiments and data, philosophy attempts to penetrate to a deeper truth through reflection and systematic thought. Its tools are logic, language and the analysis of concepts. However, it is precisely this theoretical approach that has often earned it the reputation of being dry or distant from life - yet it is the exact opposite, and indirectly focuses on the question of the good life.
“We think of ourselves as pleasure-seekers, but we are remarkably quick to call for bans and the police if anything goes against the grain of our increasingly conservative sensibilities. Our political renunciation of what we can get out of life is therefore ultimately based on an aesthetic weakness: the inability to create and appreciate the conditions under which such offensive things as partying, tobacco, alcohol, sex, black humor, idle reflection, etc. can be experienced as pleasurable. This shows that the richest populations in the world have forgotten how to ask themselves what is worth living for.” (Pfaller, 2011, What is worth living for)
So why is philosophy so important? Because it forms the foundation of our worldviews. Without philosophy, there would be no clear criteria for moral action, no idea of justice and no understanding of the principles that structure our knowledge. It is the discipline that teaches us to ask the right questions - a skill that is essential in a world full of complex challenges. Especially in the age of artificial intelligence, where technological developments are radically changing social and ethical foundations, philosophy provides orientation. It forces us to think about the implications of our actions instead of simply blindly pursuing progress. → Continue reading here.
🛰️ NEWS
Looking Forward
💰 Google Backs $90M Investment in Cassava: Google joins Cassava’s $90M funding to boost Africa’s digital infrastructure, tackling outages and expanding connectivity with subsea cables and cloud solutions.
🍎 What Apple’s AI Knows About You: Apple Intelligence delivers personalized AI features using private data—but ensures it stays secure, processed on-device or via encrypted cloud systems, never training its models.
🔢 MODELS
Cohere For AI Unveils Command R7B: A Versatile 7B-Parameter Multilingual Model
Cohere For AI has released C4AI Command R7B, a 7-billion-parameter language model optimized for advanced reasoning, summarization, code generation, and Retrieval Augmented Generation (RAG) tasks. Trained across 23 languages with a 128K context length, the model excels on benchmarks, outperforming similarly sized open models like LLaMA 3.1 and Gemma 2 IT in areas such as code, SQL, and interactive tools.
R7B supports conversational and instruct modes, tool use integration, and enhanced code translation capabilities. Available under a CC-BY-NC license, Command R7B offers researchers and developers high performance for interactive and enterprise applications. → Check it out on Hugging Face.
📽️ VIDEO
Real World AI Use Cases, Open-Source, Hallucinations (MindsDB CEO Interview)
Our recent interview with MindsDB CEO Jorge Torres delves into how MindsDB enables enterprises to transform vast, scattered datasets into actionable insights using natural language queries. Jorge shares his vision for "Enterprise AGI" as a collaborative AI system that streamlines workflows, enhances data discovery, and revolutionizes decision-making. Get the full scoop! 👇
Reply