Groq creates the world's fastest AI

PLUS: Telekom's app-less AI phone prototype

Welcome, AI enthusiasts.

Move over, GPUs — a new contender has the AI hardware space buzzing.

Groq’s new TPU tech is going viral, showcasing blazing speeds that enable near-instant chatbot responses and open the door for a new wave of innovation. Let’s explore…

In today’s AI rundown:

  • Blazing-fast Groq AI processor goes viral

  • Telecomm giant to reveal AI phone

  • Utilize Gemini across Google apps

  • AI chip uses light waves for processing

  • 6 new AI tools & 4 new AI jobs

  • More AI & tech news

Read time: 4 minutes

LATEST DEVELOPMENTS

GROQ

Image source: Groq

The Rundown: AI startup Groq’s (not Elon’s Grok…) hardware breakthroughs are going viral in the AI world, with specialized processing units that run LLMs delivering nearly instantaneous response times.

The details:

  • Groq uses LPU (language processing units) architecture instead of GPU (graphics processing unit), enabling more efficient and faster speeds.

  • Public benchmarks of Groq’s hardware hit blazing speeds of 500 tokens per second, compared to 30-50 for GPT 3.5.

  • This side-by-side demo of Groq vs. GPT 3.5 shows Groq completing the same prompt but ~4x faster.

  • Groq, founded in 2016, sent a cease-and-desist letter to Elon Musk’s X in November over its chatbot’s ‘Grok’ name reveal.

Why it matters: To understand the speeds Groq achieves, you truly have to see it to believe it. Near-instant responses open up a new world of potential for AI and user experiences, while the efficiency and affordability of LPUs could offer an alternative to in-demand GPUs on the market.

You can try Groq here using open-source LLMs like Mixtral or Llama.

TOGETHER WITH MASTERWORKS

After last year’s rate hikes, investors were finally rewarded for parking their cash with 5% APY. But for those really looking to put their money to work, there’s something else to consider: opportunity.

Like the opportunity to invest with Masterworks, the art investing platform whose 20 exits have each delivered a profit — with median returns like 14.6%, 16.4%, and 17.6%, among assets held for over one year.

Rundown readers can skip the waitlist.

Investing involves risk, and past performance is not indicative of future returns. See important Reg A disclosures and aggregate advisory performance masterworks.com/cd

DEUTSCHE TELEKOM

Image source: Deutsche Telekom

The Rundown: Deutsche Telekom, the majority stakeholder of T-Mobile, plans to release an app-free AI phone with a ‘generative interface’ called the T Phone on Feb. 26.

The details:

  • The AI phone aims to humanize interactions by replacing fixed apps with personalized experiences.

  • The company partnered with Brain.ai and Qualcomm to enable cloud-based AI to power the phone.

  • T Phone's voice-powered AI can assist with booking travel, shopping, logistics, and creative tasks.

Why it matters: The T Phone’s app-less design may fly in the face of current user-interface standards, but it could be the perfect complement to the radical changes coming with on-device AI. One thing is certain — the mobile era of artificial intelligence will enter the mainstream VERY soon.

AI TRAINING

The Rundown: Just as you can incorporate GPTs into ChatGPT chats, you can now invoke Google apps in your Gemini prompts to swiftly search and summarize content.

Step-by-step:

  1. Head to Gemini and enable Google Workspace extensions.

  2. Try prompting Gemini to summarize or locate information across your Google Workspace by simply appending @gmail, @drive, or @docs to your queries.

  3. Unlike Gmail search, Gemini can find and summarize your emails. For instance, you can ask Gemini to identify high-priority emails that require your attention.

  4. You can also ask Gemini to review a Google Doc and summarize the key points.

With Google managing your emails and documents, the possibilities are endless!

PRESENTED BY USERDOC

The Rundown: Building Software? Userdoc helps you create detailed software requirements in minutes using its interactive AI wizard that guides you through the process.

With Userdoc, you can:

  • Scope detailed requirements in 10 minutes, not 10 days

  • Get step-by-step software requirement guidance from its AI wizard

  • Collaborate with your whole team, sync with your project management tools, and export as Word, Excel, and CSV

Your software deserves to be successful. Click here to get started with Userdoc for free.

AI RESEARCH

Image source: Midjourney

The Rundown: Engineers at the University of Pennsylvania just created a first-of-its-kind AI chip that uses light waves instead of electricity for faster processing.

The details:

  • Leveraging light for computation utilizes the fastest possible means of communication.

  • The paper combines research on using light waves for processing with silicon devices.

  • The chip promises faster speed, lower energy costs, and enhanced privacy.

Why it matters: We’ve all heard the outcry over energy-hungry AI systems, as well as the vast amounts of compute needed for powering advanced models. If this light breakthrough can be applied, it could hold revolutionary potential for increasing AI capabilities.

NEW TOOLS & JOBS

  • 🏋️‍♀️ Fit Senpai- Get personalized fitness plans and eliminate gym guesswork (link)

  • 🖋️ Text Humanizer- Enhance your text with AI for human-quality writing (link)

  •  SEO by AI- Boost Google rankings with free AI-powered SEO optimization (link)

  • 👩🏻‍⚕️ DrLambda- Your AI assistant for professional content creation (link)

  • 🤖 Prompty GPT- Optimize your prompts for better results (link)

  • 🎓 Repeto- Master material with interactive learning and note-taking (link)

  • 🚀 Anyscale- Product Manager (link)

  • 🔬 Shield AI- Senior Electrical Engineer (link)

  • 🤓 Anthropic- Solutions Architect Manager (link)

  • 👩‍💻Open AI- Platform ML Engineering Manager, Inference (link)

QUICK HITS

Elon Musk weighed in on the Sora hype, saying that Tesla video exceeds OpenAI’s model in predicting ‘extremely accurate physics’ for its self-driving functionality.

Singapore announced that the country is investing over $700M into its AI capabilities over the next five years, seeking to increase access to advanced chips and setting up AI centers to push innovation.

Meta’s Yann LeCun Meta has been critical of OpenAI's Sora, arguing his JEPA architecture is better suited for building comprehensive world models for AI autonomy.

Royal Caribbean is utilizing AI on its new ‘Icon of the Seas’ mega-ship, helping the company predict food usage on-board and reduce waste.

Chinese EV maker Xpeng is reportedly investing nearly $500M into AI in 2024, with plans to also hire 4,000 new workers despite other automakers slashing costs.

The UK and its allies are exploring the use of advanced AI-enabled swarm drones to support Ukraine's defense, potentially providing a new tactical advantage against Russian forces.

THAT’S A WRAP

SPONSOR US

Get your product in front of over 450k+ AI enthusiasts

Our newsletter is read by thousands of tech professionals, investors, engineers, managers, and business owners around the world. Get in touch today.

FEEDBACK

How would you rate today's newsletter?

Vote below to help us improve the newsletter for you.

Login or Subscribe to participate in polls.

If you have specific feedback or anything interesting you’d like to share, please let us know by replying to this email.

Join the conversation

or to participate.