DeepSeek R1 and Perplexity

Hi everyone,

So much AI news happened today, that we could not get this newsletter out in time AND NVIDIA dropped by 20%. That is the single biggest market value value erosion of any company in history.

Why? DeepSeek claims it was able to train its model for mere millions instead of billions of computing power… Who know if that’s right? But if it is, it shook the US AI dominance.

The NVIDIA thing is fitting, as we’ve partnered up with a cool finance product for this issue.

Let’s get right into it.

In this issue:

• 📕Learn Something: Perplexity Assistant challenges Gemini on android

• 🤝 In Partnership: We’ve got something special for you; act fast.

• 🖼️AI Art: Examples of great and trending AI art

• 🤿Deep Dive: DeepSeek R1 sparks AI debate

• ⚒️Tool Snapshots: Tools for AI, no-code, and productivity

• 📰Top News: News on AI, no-code, and productivity

🤿 DEEP DIVE

Perplexity Assistant Launches as a New Contender in AI Capabilities for Android Users

Intelligence: Perplexity Assistant has debuted on the Google Play Store, offering Android users the ability to perform real-world tasks using apps on their devices, challenging existing AI models like Google's Gemini.

• Perplexity Assistant is now available for all Android users, positioning itself as an alternative to other AI models.

• The assistant can perform various tasks such as hailing Uber rides, making dinner reservations through OpenTable, and playing music through platforms like Spotify and YouTube.

• Unlike some existing assistants, Perplexity Assistant successfully opened Uber to find a ride, showcasing its ability to navigate outside a specific ecosystem.

• The assistant utilizes the device's camera to provide detailed analyses of objects in view, although it requires multiple attempts for accurate execution.

• Perplexity offers more nuanced interactions in certain tasks, but it is still in development, with future updates expected to enhance its capabilities.

• Perplexity Assistant is free to download, with a subscription option for $19.99/month for advanced features.

🤝IN PARTNERSHIP WITH FINANCE BUZZ

Pay No Interest Until Nearly 2027 AND Earn 5% Cash Back

Use a 0% intro APR card to pay off debt.
Transfer your balance and avoid interest charges.
This top card offers 0% APR into 2027 + 5% cash back!

Learn How To Apply Now

🖼️ AI ART

Examples of great and trending AI art

A unique series of movie posters in a journal-collage style by ordogas, made with MidJourney. These pieces mix handwritten notes, textures, and cutout images to give a fresh take on classic film visuals.

🤿 DEEP DIVE

 DeepSeek's R1 Model Sparks Controversy and Discussion in the AI Community

Intelligence: The release of DeepSeek's reasoning model R1 has generated significant attention and discussion about its implications for the AI landscape, particularly in comparison to established models from American companies.

• DeepSeek claims that R1 matches or outperforms OpenAI's o1 model in various benchmarks, drawing praise from influential figures like venture capitalist Marc Andreessen.

• The company boasts a training cost of only $5.6 million for R1, significantly lower than the hundreds of millions spent by leading American firms on AI models.

• Despite U.S. sanctions restricting advanced chip sales to Chinese firms, DeepSeek's success illustrates a push for innovation that focuses on efficiency and resource collaboration. However, the company's CEO acknowledged that these sanctions still present challenges.

• Some critics, like Curai CEO Neal Khosla, have accused DeepSeek of manipulating costs to undercut U.S. competitors, although no evidence has been provided for these claims.

• Analysts like journalist Holger Zschaepitz warn that DeepSeek's advancements could challenge the viability of U.S. investments in AI, while others, like Y Combinator CEO Garry Tan, suggest that lower training costs could increase the demand for AI applications.

• Meta's Chief AI Scientist Yann LeCun highlighted the importance of open-source research, suggesting that DeepSeek’s innovations build on the work of others, with benefits extending beyond national boundaries.

⚒️ TOOL SNAPSHOTS

Futuristic tools within AI, no-code, and productivity

• 💻 Trae - Enhancing workflow speed with adaptive AI collaboration. Free to use.

• 📁 DROP - Share and host large files securely and professionally. Free option available.

• 💡 Gemini 2.0 Flash Thinking - Google's advanced, explorable reasoning model. Free to try.

• 🤖 Taskek - Your AI-driven solution for all-in-one team collaboration. Free to use.

📰 TOP NEWS

News on AI, no-code, automation, and productivity

⚖️ Character.ai Faces Lawsuit Over Chatbot Safety

Character.ai seeks dismissal of a wrongful death lawsuit alleging its chatbots contributed to a teen's suicide, highlighting safety measures added since the incident.

🧪 Humanity's Last Exam Tests AI's Limits

A new benchmark from CAIS and Scale AI evaluated advanced AI models on world-class expert-level reasoning, revealing they could answer fewer than 10% of the toughest questions correctly. The initiative aims to identify gaps and guide future AI research.

🌐 OGOpenAI Redirects to DeepSeek

A software engineer purchased "OGOpenAI.com" and redirected it to DeepSeek, a Chinese AI lab gaining recognition for open-source AI models that are free and offline-accessible, echoing OpenAI's early approach.

🤖 Galileo Launches Agentic Evaluations

Galileo introduces a tool to ensure AI agents work reliably, helping enterprises monitor performance, detect errors, and manage costs as they scale AI deployments.

ℹ️ ABOUT US

The Intelligent Worker helps you to be more productive at work with AI, automation, no-code, and other technologies.

We like real, practical, and tangible use-cases and hate hand-wavy, theoretical, and abstract concepts that don’t drive real-world outcomes.

Our mission is to empower individuals, boost their productivity, and future-proof their careers.

We read all your comments - please provide your feedback!